Download ebook New Perspectives On Enterprise Decision Making Applying Artificial Intelligence Techniques 1St Edition Julian Andres Zapata Cortes Editor Giner Alor Hernandez Editor Cuauhtemoc Sanchez Ramirez Editor online pdf all chapter docx epub

New Perspectives on Enterprise
Decision Making Applying Artificial

Intelligence Techniques 1st Edition
Julian Andres Zapata Cortes Editor
Giner Alor Hernández Editor
Cuauhtémoc Sánchez Ramírez Editor
Jorge Luis García Alcaraz Editor
Visit to download the full and correct content document:
https://ebookmeta.com/product/new-perspectives-on-enterprise-decision-making-appl
ying-artificial-intelligence-techniques-1st-edition-julian-andres-zapata-cortes-editor-gin
er-alor-hernandez-editor-cuauhtemoc-sanchez-ramirez-editor/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...
Handbook on Decision Making Volume 3 Trends and

Challenges in Intelligent Decision Support Systems 1st
Edition Julian Andres Zapata Cortes Cuauhtémoc Sánchez
Ramírez Giner Alor Hernández Jorge Luis García Alcaraz
https://ebookmeta.com/product/handbook-on-decision-making-
volume-3-trends-and-challenges-in-intelligent-decision-support-
systems-1st-edition-julian-andres-zapata-cortes-cuauhtemoc-
sanchez-ramirez-giner-alor-hernandez-jorge-luis-ga/
Chess Explained The c3 Sicilian 1st Edition Sam Collins
https://ebookmeta.com/product/chess-explained-
the-c3-sicilian-1st-edition-sam-collins/
Starting Out The c3 Sicilian 1st Edition John Emms
https://ebookmeta.com/product/starting-out-the-c3-sicilian-1st-
edition-john-emms/
The Paper Issue 83 1st Edition Origamiusa
https://ebookmeta.com/product/the-paper-issue-83-1st-edition-
origamiusa/
The American Revolution 1774 83 2nd Edition Daniel
Marston
https://ebookmeta.com/product/the-american-
revolution-1774-83-2nd-edition-daniel-marston/
BNF 83 (British National Formulary) March 2022 Joint

Formulary Committee
https://ebookmeta.com/product/bnf-83-british-national-formulary-
march-2022-joint-formulary-committee/
English Grammar Exercises with answers Part 4 Your

quest towards C2 1st Edition Daniel B. Smith
https://ebookmeta.com/product/english-grammar-exercises-with-
answers-part-4-your-quest-towards-c2-1st-edition-daniel-b-smith/
Patriot vs Loyalist American Revolution 1775 83 Combat

1st Edition Si Sheppard
https://ebookmeta.com/product/patriot-vs-loyalist-american-
revolution-1775-83-combat-1st-edition-si-sheppard/
British Light Infantryman vs Patriot Rifleman American

Revolution 1775 83 1st Edition Robbie Macniven
https://ebookmeta.com/product/british-light-infantryman-vs-
patriot-rifleman-american-revolution-1775-83-1st-edition-robbie-
macniven/
Studies in Computational Intelligence 966
Julian Andres Zapata-Cortes

Giner Alor-Hernández
Cuauhtémoc Sánchez-Ramírez
Jorge Luis García-Alcaraz Editors
New Perspectives
on Enterprise
Decision-Making
Applying Artificial
Intelligence
Techniques
Studies in Computational Intelligence
Volume 966
Series Editor
Janusz Kacprzyk, Polish Academy of Sciences, Warsaw, Poland
The series “Studies in Computational Intelligence” (SCI) publishes new develop-
ments and advances in the various areas of computational intelligence—quickly and
with a high quality. The intent is to cover the theory, applications, and design
methods of computational intelligence, as embedded in the fields of engineering,
computer science, physics and life sciences, as well as the methodologies behind
them. The series contains monographs, lecture notes and edited volumes in
computational intelligence spanning the areas of neural networks, connectionist
systems, genetic algorithms, evolutionary computation, artificial intelligence,
cellular automata, self-organizing systems, soft computing, fuzzy systems, and
hybrid intelligent systems. Of particular value to both the contributors and the
readership are the short publication timeframe and the world-wide distribution,
which enable both wide and rapid dissemination of research output.
Indexed by SCOPUS, DBLP, WTI Frankfurt eG, zbMATH, SCImago.
All books published in the series are submitted for consideration in Web of Science.
More information about this series at http://www.springer.com/series/7092

Julian Andres Zapata-Cortes ·
Giner Alor-Hernández ·
Cuauhtémoc Sánchez-Ramírez ·
Jorge Luis García-Alcaraz
Editors
New Perspectives
on Enterprise
Decision-Making Applying
Artificial Intelligence
Techniques
Editors
Julian Andres Zapata-Cortes Giner Alor-Hernández
CEIPA Business School Division of Research and Postgraduate
Fundación Universitaria CEIPA Studies
Sabaneta, Colombia Instituto Tecnológico de Orizaba
Tecnológico Nacional de México
Cuauhtémoc Sánchez-Ramírez Orizaba, Mexico
Division of Research and Postgraduate
Studies Jorge Luis García-Alcaraz
Instituto Tecnológico de Orizaba Autonomous University of Ciudad Juarez
Tecnológico Nacional de México Ciudad Juárez, Chihuahua, Mexico
Orizaba, Mexico
Division of Research and Postgraduate
Studies
Tecnológico Nacional de México/Instituto
Tecnológico de Ciudad Juárez
Ciudad Juárez, Chihuahua, Mexico
ISSN 1860-949X ISSN 1860-9503 (electronic)

Studies in Computational Intelligence
ISBN 978-3-030-71114-6 ISBN 978-3-030-71115-3 (eBook)
https://doi.org/10.1007/978-3-030-71115-3
© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature
Switzerland AG 2021
This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether
the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse
of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and
transmission or information storage and retrieval, electronic adaptation, computer software, or by similar
or dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this book
are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or
the editors give a warranty, expressed or implied, with respect to the material contained herein or for any
errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional
claims in published maps and institutional affiliations.
This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Preface
With the increase in computing power produced in the last decade, the ability to
analyze large datasets and the development of new algorithms have produced the
possibility to develop mechanisms that allow decision-making processes that in past
years were thought impossible to be carried out. These mechanisms have the capacity
to surpass the human capacity for these tasks with the ability to carry out the anal-
ysis autonomously and to learn from themselves, allowing them to be conceived as
artificial intelligence.
Artificial Intelligence (AI) techniques cover the automation of cognitive and phys-
ical tasks. These techniques help people to perform tasks faster and make better deci-
sions. It enables the automation of decision-making process without human inter-
vention. AI techniques can enhance automation by reducing intensive human labor
and tedious tasks. There are more forms in which Artificial Intelligence is making
a difference for enterprises in decision-making marketing, customer relationship
management, recommender systems, problem-solving, opinion mining, augmented
analytics, to mention but a few.
In marketing, it is necessary to understand customer needs and desires and aligning
products to those needs and desires. A handle on changing customer behavior is vital
to make the best marketing decisions. AI simulation and modeling techniques provide
reliable insight into the consumers’ persona. This will help predicting consumers’
behavior. Through real-time data gathering, trend analysis, and forecasting, an AI
system can help businesses make insightful marketing decisions. Furthermore, orga-
nizations can identify a consumer’s lifetime value with the help of AI’s buyer persona
modeling. It can help organizations manage multiple inputs. During a complex
decision-making process, AI can efficiently manage and control different factors
at the same point in time. It can source and process large amounts of data within
minutes while providing valuable business-based insights. While we humans face
decision fatigue, algorithms do not have such limitations, which make AI-based
decisions faster and better. Also, AI techniques have provided businesses invalu-
able insight about consumers, which helps them enhance their communication with
the consumers. It also helps retailers predict product demand and respond to it
quickly. For this end, opinion mining helps businesses understand why people feel
v
vi Preface
the way they feel. Most often a single customer’s concerns might be common among
others. When sufficient opinions are gathered and analyzed correctly, the information
gleaned will help organizations gauge and predict the concerns of the silent majority.
AI has improved this mining process through automation, which is quicker and more
reliable, helping organizations in making critical business decisions.
In e-commerce, an AI system learns a consumer’s preference based on “explicit”
or “implicit” feedbacks. This kind of systems are called recommender systems. A
recommender system can provide information helping the organization to reduce
bounce rate and craft better customer-specific targeted content. Wise business deci-
sions are made when business executives and decision-makers have reliable data
and recommendations. AI not only improves the performance of both the individual
members of the team but also the competitive edge of the business.
In medicine, the artificial intelligence allows the improvement of many tasks,
ranging from better management of resources such as rooms and personnel, as well
as achieving more accurate and faster medical diagnoses, such as computer-assisted
radiologies or the use of robots to perform less invasive surgeries. In the service
industry, AI is used in automated online assistants that allow faster and lower cost
service, which has the ability to learn from past interactions, thus improving customer
service.
Based on this understanding, there are many other applications of artificial intelli-
gence in many knowledge fields, and this is the main reason and the core of the present
book, which is aimed to disseminate current trends among innovative and high-
quality research regarding the implementation of conceptual frameworks, strate-
gies, techniques, methodologies, informatics platforms, and models about Enter-
prise Decision-Making Applying Artificial Intelligence Techniques. The specific
objectives of the book can be summarized as follows:
• Generate a dissemination venue from both the academia and the industry in the
topics studied in the book, presenting cases of new approaches, applications,
methods, and techniques of the application of Artificial Intelligence in enterprise
decision-making.
• Generate a collection of theoretical and practical research works in the field of
Artificial Intelligence Techniques applied in enterprise decision-making.
• Establish the state of the art in the field of Artificial Intelligence Techniques
applied in enterprise decision-making.
This book is composed of a set of chapters, each of the kind of regular research
papers. These works have been edited according to the norms and guidelines of
Springer Verlag Editorial. Several calls for chapters were distributed among the
main mailing lists of the field for researchers to submit their works to this issue. 25
expressions of interest in the form of abstracts were received in total, which were
subject to a screening process to ensure their clarity, authenticity, and relevancy to
this book. These proposals came from several countries such as Colombia, Mexico,
Spain, Perú, and Ukraine.
After the abstract reviewing process, 25 proposals were accepted and asked to
submit full versions. These versions were reviewed by at least two pairs in order to
Preface vii
ensure the relevance and quality of the documents. After this process, 21 chapters
were finally accepted for their publication once the corrections requested by the pairs
and the editors were completed.
The book content is structured in three parts: (1) Industrial Applications, (2)
Decision-Making Systems for Industry, and (3) Artificial Intelligence Techniques.
The chapters in each of these parts are as follows.
Part I Industrial Applications: This part contains seven chapters.
Chapter 1, entitled “Merging Event Logs for Inter-organizational Process Mining,”
presents a methodology to merge the event logs of the different partners of a collab-
orative business process, in order to serve as input for the process mining algorithm.
On the one hand, the methodology consists of a method set for searching the corre-
lation between events of the log of different partners involved in the collaboration.
These methods are implemented at the trace level and the activity level. On the other
hand, the methodology consists of a set of methods and rules for discovering process
choreography. From the knowledge gained by the above methods, message-type
tasks are identified and marked in each event log, and then using a formal set of
rules, the message task sub-type (send or receive) is discovered. Finally, links using
message sequence flow connectors between message tasks identified as pair activi-
ties in event logs are automatically defined. The proposed approach is tested using
a real-life event log that confirms their effectiveness and efficiency in the automatic
specification of message flows of the process choreography discovered, allowing to
build a collaborative business process model.
Chapter 2, entitled “Towards Association Rule-Based Item Selection Strategy
in Computerized Adaptive Testing,” proposes the integration of Association Rule
Mining as an item selection criterion in a CAT system. Specifically, we present the
analysis of association rule mining algorithms such as Apriori, FPGrowth, Predic-
tiveApriori, and Tertius into three datasets obtained from the subject databases to
know the advantages and disadvantages of each algorithm and choose the most suit-
able one to employ in an association rule-based CAT system that is being developed as
a Ph.D. project. We compare the algorithms considering the number of rules discov-
ered, average support and confidence, lift, and velocity. According to the experiments,
Apriori found rules with greater confidence, support, lift, and in less time.
Chapter 3, entitled “Uncertainty Linguistic Summarizer to Evaluate the Perfor-
mance of Investment Funds,” proposes a methodology to implement the uncertain
linguistic summarizer posed in Liu’s uncertain logic to measure the performance of
investment funds in the Colombian capital market. The algorithm extracts a truth
value for a set of linguistic summaries, written as propositions in predicate logic,
where the terms for the quantifier, subject, and predicate are unsharp. The linguistic
summarizer proves to be autonomous, successful, efficient, and close to human
language. Furthermore, the implementation has a general scope and could become a
data mining tool under uncertainty. The propositions found characterize with plenty
of sense the investment funds data. Finally, a corollary that allows accelerating the
obtention of the summaries is presented.
viii Preface
Chapter 4, entitled “Map-Bot: Mapping Model of Indoor Work Environments in

Mobile Robotics,” presents a mapping model of indoor work environments in mobile
robotics, called Map-Bot. The model integrates hardware and software modules
for navigation, data acquisition and transfer and mapping. Additionally, the model
incorporates a computer that runs the software responsible for the construction of two-
dimensional representations of the environment (Vespucci module), a mobile robot
that collects sensory information from the workplace and a wireless communication
module for data transfer between the computer and the robot. The results obtained
allow the implementation of the reactive behavior “follow walls” located on its right
side on paths of 560 cm. The model allowed to reach a safe and stable navigation for
indoor work environments using this distributed approach.
Chapter 5, entitled “Production Analysis of the Beekeeping Chain in Vichada,
Colombia. A System Dynamics Approach,” presents a model of beekeeping produc-
tion in the region of Vichada in Colombia. The beekeeping chain was chosen because
it is a sector of great economic importance in the mentioned region which has
the highest indices of multidimensional poverty in Colombia, but also it is one of
the places with the greatest conservation of its biodiversity. A systems dynamics
approach is used from a causal diagram to explain the interactions among bee rearing,
wax production, honey production, and transformation, and then simulations were
performed to determine the behavior of inventories concerning the production and
demand. This model highlights the dynamics of the system and the management of the
supply chain and is presented as a useful tool to predict production-demand scenarios
in the beekeeping sector where similar studies are scarce. As future research, it is
recommended to include the economic nature of the products in this kind of models
so that scenarios can be proposed to help beekeepers make production decisions
according to demand, and develop inventory policies.
Chapter 6, entitled “Effect of TPM and OEE on the Social Performance of Compa-
nies,” reports a model of structural equations that integrates three independent vari-
ables: Total productive maintenance, Just in time and Overall equipment efficiency,
and the relationship they have with Social sustainability as a dependent variable. The
four variables are related through six hypotheses that are validated with informa-
tion gathered from 239 questionnaires answered by executives laboring at Mexican
maquiladora industry. The partial least squares technique is used to statistically vali-
date the relationships among variables. Findings indicate that Total predictive main-
tenance has a strong impact on Overall equipment efficiency and Just in time, and
the variables that most influence Social sustainability are Total predictive mainte-
nance and Just in time. It is concluded that Social sustainability can be obtained
through proper use and maintenance of the machines and with timely fulfillment of
production orders.
Chapter 7, entitled “ENERMONGRID: Intelligent Energy Monitoring, Visual-
ization and Fraud Detection for Smart Grids,” presents a tool used for intelligent
energy monitoring, data visualization, and fraud detection in electric networks to
generate rich information in near real time, which can be used to make decisions
for the optimal energy production, generation, distribution, and consumption. This
tool allows solving many problems that can arise when dealing with energy load
Preface ix
estimates, loss estimates, as well as fraud detection and prevention, in the entities in
charge of managing an electric network, with the focus of doing so in smart grids.
Part II Decision-Making Systems for Industry: This part contains seven chapters.
Chapter 8, entitled “Measuring Violence Levels in Mexico Through Tweets,”
proposes a novel way to evaluate what people say and how they feel about violence
by analyzing Twitter data. To do this, we describe a methodology to create an indi-
cator to denote social perception. This methodology uses technologies like Big Data,
Twitter analytics, Web mining, and Semantic Web by manipulating software like
ELK (Elasticsearch for data storage, Logstash for collecting data, and Kibana for
data visualization); SPSS and R for statistical data analysis; and Atlas.ti, Ghephi,
and Wordle for semantic analysis. At the end of the chapter, we show our results
with a word cloud, a social graph, and the indicator of social perception of violence
in Mexico at the federal entity and metropolitan zone levels.
Chapter 9, entitled “Technology Transfer from a Tacit Knowledge Conserva-
tion Model into Explicit Knowledge in the Field of Data Envelopment Analysis,”
presents a preservation model of production engineers’ tacit knowledge of Data
Envelopment Analysis (DEA). Their expertise was explicitly coded into a computer
system, and the model was developed by applying techniques and procedures from
the fields of engineering and knowledge management. Technology transfer enables to
solve the problem of selecting criteria and interpreting results with DEA techniques,
when the efficiency of similar organizations is compared using an efficient frontier
derived from non-parametric approximations of such techniques. Misunderstanding
the techniques leads to misinterpretations of DEA results. This model was created
by applying Knowledge Engineering, which enables to preserve and extend specific
experiences and expertise in time by means of computer solutions. The model had an
efficient and positive impact on strategic self-learning processes for the community
interested in production engineering, knowledge transfer, and management.
Chapter 10, entitled “Performance Analysis of Decision Aid Mechanisms for
Hardware Bots Based on ELECTRE III and Compensatory Fuzzy Logic,” proposes
two novel cognitive preference models viable for hardware with few memory cells
and small processing capacity. These are the Outranking Relations (or OR) and
Compensatory Fuzzy Logic (or CFL) which are two techniques that lack a study of
their hardware performance as intelligence modelers, and despite their simple defini-
tion and generalization capacity, hardware agents almost never use them. The chapter
analyzes the feasibility of implementing Outranking Relations (OR) and Compen-
satory Fuzzy Logic (CFL) in hardware platforms with low resources, highlighting
that competitiveness of the proposed tools and the arising of new research lines.
Chapter 11, entitled “A Brief Review of Performance and Interpretability in Fuzzy
Inference Systems,” presents a scoping review related to fuzzy logic emphasizing
in Compensatory Fuzzy Logic (CFL), Archimedean Compensatory Fuzzy Logic
(ACFL), and inference systems. It presents the literature analysis on surveys and
general reviews to contrast the scoping review. The chapter also presents the research
analysis through a case study, a comparison of compensatory fuzzy logic with other
fuzzy logic structures and other related works.
x Preface
Chapter 12, entitled “Quality and Human Resources, Two JIT Critical Success
Factors,” presents a structural equation model associating JIT elements of quality
planning and quality management with human resources and economic performance
in the context of Mexican maquiladoras (cross-border assembly plants). Results
indicate that even though there is no direct relationship between quality planning and
management and economic benefits, these variables are indirectly related through
human resources. In conclusion, human resources are key to achieving financial
success of maquiladoras.
Chapter 13, entitled “Operational Risks Management in the Reverse Logistics of
Lead-Acid Batteries,” develops an operational risk identification and prioritization in
the reverse logistics of lead-acid batteries in Colombia using the questionnaires and
FQFD (fuzzy quality function deployment) approaches. We identified operational
risks in the reverse logistics of lead batteries and a probability-impact matrix to define
which of these risks should be prioritized with FQFD. In this way, we established
the priority of the risks considered. These prioritized risks help organizations related
to this activity to develop action plans to mitigate these risks. Once the most critical
risks are defined, actions to mitigate or eliminate them were proposed.
Chapter 14, entitled “Dynamic Evaluation of the Livestock Feed Supply Chain
from the Use of Ethanol Vinasses,” proposes a conceptual design of the vinasse-based
livestock feed supply chain using system dynamics to identify the key variables of
the chain and assess how vinasse can be efficiently used to produce animal feed
and other products such as ethanol. Both industrial and scientific efforts are made to
find alternative uses for vinasse in energy generation, soil fertilization, and livestock
feed production. However, this last alternative has not been sufficiently explored.
The results demonstrate that a continuous supply of molasses ensures a contin-
uous production of ethanol, which in turn guarantees constant vinasse availability to
produce livestock feed.
Part III Artificial Intelligence Techniques: This part contains seven chapters.
Chapter 15, entitled “Comparative Analysis of Decision Tree Algorithms for Data
Warehouse Fragmentation,” presents the analysis of different decision tree algo-
rithms to select the best one to implement the fragmentation method. Such analysis
was performed under version 3.9.4 of Weka considering four evaluation metrics
(precision, ROC area, recall, and F-measure) for different selected datasets using
the SSB (Star Schema Benchmark). Several experiments were carried out using two
attribute selection methods, Best First and Greedy Stepwise, the datasets were pre-
processed using the Class Conditional Probabilities filter, and the analysis of two
datasets (24 and 50 queries) with this filter was included to know the behavior of
the decision tree algorithms for each dataset. Once the analysis was concluded, we
can determine that for 24 queries dataset the best algorithm was RandomTree since
it won in two methods. On the other hand, in the dataset of 50 queries, the best deci-
sion tree algorithms were LMT and RandomForest because they obtained the best
performance for all methods tested. Finally, J48 was the selected algorithm when
neither an attribute selection method nor the class probabilities filter is used. But, if
Preface xi
only the latter is applied to the dataset, the best performance is given by the LMT
algorithm.
Chapter 16, entitled “Data Analytics in Financial Portfolio Recovery Manage-
ment,” presents the application of data analytics and machine learning techniques
to predict the behavior of the loan default in a non-financial entity. Five classifica-
tion algorithms (neural networks, decision trees, support vector machines, logistic
regression, and K neighbors) were run on a dataset of credit behavior data. Decision
trees have shown the best prediction performance to determine whether a loan will
be paid or become irrecoverable after running five predictive models.
Chapter 17, entitled “Task Thesaurus as a Tool for Modeling of User Information
Needs,” describes some examples of task thesaurus usage in intelligent applica-
tions for adaptation to user needs. Task thesaurus is an element of user model that
reflects dynamic aspects of user’s current work. Such thesaurus is based on domain
ontology and contains the subset of its concepts that are dealt with user task. These
can be generated automatically by analysis of task description or with the help of
semantic similarity estimations of ontological concepts. Task thesaurus represents
personalized user view on domain and depends on his/her abilities, experience, and
aims.
Chapter 18, entitled “NHC_MDynamics: High-Throughput Tools for Simulations
of Complex Fluids Using Nosé-Hoover Chains and Big Data Analytics,” is focused
on the implementation of the NVT algorithm of molecular dynamics based on the
Nosé-Hoover thermostat chains with the high-performance computing technology
such as Graphical Processing Units (GPUs) and Big Data analytics for the generation
of knowledge to help understand the functioning of thermodynamic properties in
simulated systems of Lennard-Jones fluids, as well as the study of the behavior of
proteins, such as diabetes, to refine their structures to formulate and integrate them
into the improvement of food in an improved diet for people with this condition.
Chapter 19, entitled “Determination of Competitive Management Perception
in Family Business Leaders Using Data Mining,” seeks to determine competitive
management perception of family business leaders, in order to establish working
assumptions in new research and propose improvement and consolidation initiatives
for these types of companies. This non-probabilistic, intentional study applied an
instrument with 10 dimensions and 94 variables to a sample of 133 family business
leaders from an intermediate city and a large city in Colombia. Data collection was
achieved using supervised machine learning algorithms in the Python programming
language, using techniques such as Cronbach’s Alpha Test, KMO, Levene, Bartlett,
Discriminant Analysis, and Decision Trees. The results allow us to identify four main
components in 19 variables: Management and technology, Quality Management,
Compensation, and Country competitiveness.
Chapter 20, entitled “A Genetic Algorithm for Solving the Inventory Routing
Problem with Time Windows,” presents a genetic algorithm that allows to simultane-
ously optimize inventory allocation and transport routes to supply a set of customers
for a specific time horizon. This model allows to obtain a minimum total cost as a
result of a better combination of the inventories at customers’ facilities and the trans-
portation required to supply them. The proposed model and the algorithm developed
xii Preface
for its solution allowed to obtain significant savings compared to the routing opti-
mization to supply all customers in each period using the vehicle routing problem
model with time windows, which allows to optimize customers’ inventory and mini-
mize transport costs in each period. However, when this solution is compared with
the total distribution cost throughout the time horizon, it generates higher costs than
the solution generated by the IRP with time windows presented in this work.
Finally, Chap. 21, entitled “Emotion Detection from Text in Learning Environ-
ments: A Review,” introduces a literature review of text-based emotion detection
in learning environments. We analyze the main APIs and tools available today for
emotion detection and discuss their key characteristics. Also, we introduce a case
study to detect the positive and negative polarities of two educational resources to
identify the accuracy of the results obtained from five selected APIs. Finally, we
discuss our conclusions and suggestions for future work.
Once a brief summary of chapters has been provided, the editors would like to
express their gratitude to the reviewers who kindly accepted to contribute in the
chapters’ evaluation at all stages of the editing process.
Medellín, Colombia Julian Andres Zapata-Cortes

Orizaba, Mexico Giner Alor-Hernández
Orizaba, Mexico Cuauhtémoc Sánchez-Ramírez
Ciudad Juárez, Mexico Jorge Luis García-Alcaraz
Acknowledgements
This book is part of the effort of several organizations that support research as the
CEIPA Business School, National Council of Science and Technology in Mexico
(CONACYT), PRODEP, Tecnológico Nacional de Mexico/IT Orizaba, and the
Autonomous University of Ciudad Juarez (UACJ).
Also, we, the editors, want to express our gratitude to other organizations that share
the same purpose but are not mentioned here due to the long list. We also appreciate
and recognize the trust, effort, and collaboration and patience that all authors and
co-workers endorse us as editors. Finally, we are thankful and appreciate the Springer
Publishing experts, especially to Thomas Ditzinger for his invaluable support at any
moment, for providing good tips for editing from their experience, patience, and
advice to materialize this book.
xiii
Contents
Part I Industrial Applications

1 Merging Event Logs for Inter-organizational Process Mining . . . . . . 3
Jaciel David Hernandez-Resendiz, Edgar Tello-Leal,
Heidy Marisol Marin-Castro, Ulises Manuel Ramirez-Alcocer,
and Jonathan Alfonso Mata-Torres
2 Towards Association Rule-Based Item Selection Strategy
in Computerized Adaptive Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Josué Pacheco-Ortiz, Lisbeth Rodríguez-Mazahua,
Jezreel Mejía-Miranda, Isaac Machorro-Cano,
and Ulises Juárez-Martínez
3 Uncertainty Linguistic Summarizer to Evaluate
the Performance of Investment Funds . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
Carlos Alexander Grajales and Santiago Medina Hurtado
4 Map-Bot: Mapping Model of Indoor Work Environments
in Mobile Robotics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
Gustavo Alonso Acosta-Amaya, Andrés Felipe Acosta-Gil,
Julián López-Velásquez, and Jovani Alberto Jiménez-Builes
5 Production Analysis of the Beekeeping Chain in Vichada,
Colombia. A System Dynamics Approach . . . . . . . . . . . . . . . . . . . . . . . 97
Lizeth Castro-Mercado, Juan Carlos Osorio-Gómez,
and Juan José Bravo-Bastidas
6 Effect of TPM and OEE on the Social Performance
of Companies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
Adrián Salvador Morales-García, José Roberto Díaz-Reza,
and Jorge Luis García-Alcaraz
xv
xvi Contents
7 ENERMONGRID: Intelligent Energy Monitoring,

Visualization and Fraud Detection for Smart Grids . . . . . . . . . . . . . . . 143
Miguel Lagares-Lemos, Yuliana Perez-Gallardo,
Angel Lagares-Lemos, and Juan Miguel Gómez-Berbís
Part II Decision-Making Systems for Industry

8 Measuring Violence Levels in Mexico Through Tweets . . . . . . . . . . . . 169
Manuel Suárez-Gutiérrez, José Luis Sánchez-Cervantes,
Mario Andrés Paredes-Valverde, Erasto Alfonso Marín-Lozano,
Héctor Guzmán-Coutiño, and Luis Rolando Guarneros-Nolasco
9 Technology Transfer from a Tacit Knowledge Conservation
Model into Explicit Knowledge in the Field of Data
Envelopment Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
Diana María Montoya-Quintero, Olga Lucia Larrea-Serna,
and Jovani Alberto Jiménez-Builes
10 Performance Analysis of Decision Aid Mechanisms
for Hardware Bots Based on ELECTRE III and Compensatory
Fuzzy Logic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
Claudia Castillo-Ramírez, Nelson Rangel-Valdez,
Claudia Gómez-Santillán, M. Lucila Morales-Rodríguez,
Laura Cruz-Reyes, and Héctor J. Fraire-Huacuja
11 A Brief Review of Performance and Interpretability in Fuzzy
Inference Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
José Fernando Padrón-Tristán, Laura Cruz-Reyes,
Rafael Alejandro Espín-Andrade, and Carlos Eric Llorente-Peralta
12 Quality and Human Resources, Two JIT Critical Success
Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
Jorge Luis García-Alcaraz, José Luis Rodríguez-Álvarez,
Jesús Alfonso Gil-López, Mara Luzia Matavelli de Araujo,
and Roberto Díaz-Reza
13 Operational Risks Management in the Reverse Logistics
of Lead-Acid Batteries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289
Daniela Sarria-Cruz, Fabio Andrés Álvarez-López,
Carolina Lima-Rivera, and Juan Carlos Osorio-Gómez
14 Dynamic Evaluation of Livestock Feed Supply Chain
from the Use of Ethanol Vinasses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309
Rocío Ramos-Hernández, Cuauhtémoc Sánchez-Ramírez,
Yara Anahí Jiménez-Nieto, Adolfo Rodríguez-Parada,
Martín Mancilla-Gómez, and Juan Carlos Nuñez-Dorantes
Contents xvii
Part III Artificial Intelligence Techniques

15 Comparative Analysis of Decision Tree Algorithms for Data
Warehouse Fragmentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 337
Nidia Rodríguez-Mazahua, Lisbeth Rodríguez-Mazahua,
Asdrúbal López-Chau, Giner Alor-Hernández,
and S. Gustavo Peláez-Camarena
16 Data Analytics in Financial Portfolio Recovery Management . . . . . . 365
Jonathan Steven Herrera Román, John W. Branch,
and Martin Darío Arango-Serna
17 Task Thesaurus as a Tool for Modeling of User Information
Needs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 385
J. Rogushina and A. Gladun
18 NHC_MDynamics: High-Throughput Tools for Simulations
of Complex Fluids Using Nosé-Hoover Chains and Big Data
Analytics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 405
Luis Rolando Guarneros-Nolasco, Manuel Suárez-Gutiérrez,
Jorge Mulia-Rodríguez, Roberto López-Rendón,
Francisco Villanueva-Mejía, and José Luis Sánchez-Cervantes
19 Determination of Competitive Management Perception
in Family Business Leaders Using Data Mining . . . . . . . . . . . . . . . . . . 435
Ángel Rodrigo Vélez-Bedoya, Liliana Adriana Mendoza-Saboyá,
and Jenny Lorena Luna-Eraso
20 A Genetic Algorithm for Solving the Inventory Routing
Problem with Time Windows . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 463
Julian Andres Zapata-Cortes, Martin Darío Arango-Serna,
Conrado Augusto Serna-Úran, and Hermenegildo Gil-Gómez
21 Emotion Detection from Text in Learning Environments:
A Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 483
Maritza Bustos-López, Nicandro Cruz-Ramírez,
Alejandro Guerra-Hernández, Laura Nely Sánchez-Morales,
and Giner Alor-Hernández
Contributors
Gustavo Alonso Acosta-Amaya Facultad de Ingeniería, Departamento de Instru-

mentación y Control, Politecnico Colombiano, Medellín, Antioquia, Colombia
Andrés Felipe Acosta-Gil Facultad de Minas, Departamento de Ciencias de la
Computación y de la Decisión, Universi-dad Nacional de Colombia, Medellín,
Antioquia, Colombia
Giner Alor-Hernández Tecnológico Nacional de México/ IT Orizaba, Orizaba,
Veracruz, Mexico
Fabio Andrés Álvarez-López Escuela de Ingeniería Industrial, Universidad del
Valle, Cali, Colombia
Martin Darío Arango-Serna Facultad de Minas, Universidad Nacional de
Colombia, Medellín, Antioquia, Colombia
John W. Branch Universidad Nacional de Colombia, Medellín, Colombia
Juan José Bravo-Bastidas Valle del Cauca, Escuela de Ingeniería Industrial,
Logistic and Production Research Group, Universidad del Valle, Cali, Colombia
Maritza Bustos-López Centro de Investigación en Inteligencia Artificial, Univer-
sidad Veracruzana, Xalapa, México
Claudia Castillo-Ramírez Tecnológico Nacional de México, Instituto Tecnológico
de Ciudad Madero, Ciudad Madero, Tamaulipas, Mexico
Lizeth Castro-Mercado Valle del Cauca, Escuela de Ingeniería Industrial, Logistic
and Production Research Group, Universidad del Valle, Cali, Colombia
Nicandro Cruz-Ramírez Centro de Investigación en Inteligencia Artificial,
Universidad Veracruzana, Xalapa, México
Laura Cruz-Reyes Tecnológico Nacional de México, Instituto Tecnológico de
Ciudad Madero, Ciudad Madero, Tamaulipas, Mexico
Mara Luzia Matavelli de Araujo Department of Business and Economy, Univer-
sity of La Rioja, Logroño, Spain
xix
xx Contributors
Roberto Díaz-Reza Department of Electric Engineering and Computation, Univer-

sidad Autónoma de Ciudad Juárez, Chihuahua, México
José Roberto Díaz-Reza Department of Electric Engineering and Computation,
Universidad Autónoma de Ciudad Juárez, Ciudad Juárez, Chihuahua, Mexico
Rafael Alejandro Espín-Andrade Autonomous University of Coahuila, Saltillo,
Mexico
Héctor J. Fraire-Huacuja Tecnológico Nacional de México, Instituto Tecnológico
Jorge Luis García-Alcaraz Department of Industrial Engineering and Manufac-
turing, Universidad Autónoma de Ciudad Juárez, Ciudad Juárez, Chihuahua,
Mexico
Hermenegildo Gil-Gómez Universidad Politécnica de Valencia, Valencia, España
Jesús Alfonso Gil-López Department of Business and Economy, University of La
Rioja, Logroño, Spain
A. Gladun International Research and Training Center of Information Technologies
and Systems of National Academy of Sciences of Ukraine and Ministry of Education
and Science of Ukraine, Kyiv, Ukraine
Juan Miguel Gómez-Berbís Department of Computer Science, Universidad
Carlos III de Madrid, Madrid, Spain
Claudia Gómez-Santillán Tecnológico Nacional de México, Instituto Tecnológico
Carlos Alexander Grajales Universidad de Antioquia, Medellín, Colombia
Luis Rolando Guarneros-Nolasco Tecnológico Nacional de México/I. T. Orizaba,
Orizaba, Veracruz, Mexico
Alejandro Guerra-Hernández Centro de Investigación en Inteligencia Artificial,
Universidad Veracruzana, Xalapa, México
Héctor Guzmán-Coutiño Universidad Veracruzana, Xalapa, Veracruz, México
Jaciel David Hernandez-Resendiz Unidad Académica Multidisciplinaria
Reynosa-RODHE, Universidad Autónoma de Tamaulipas, Reynosa, Tamaulipas,
México
Jonathan Steven Herrera Román Universidad Nacional de Colombia, Medellín,
Colombia
Jovani Alberto Jiménez-Builes Facultad de Minas, Departamento de Ciencias de
la Computación y de la Decisión, Universi-dad Nacional de Colombia, Medellín,
Antioquia, Colombia
Contributors xxi
Yara Anahí Jiménez-Nieto Faculty of Accounting and Administration, Univer-

sidad Veracruzana Campus Ixtaczoquitlán, Ixtaczoquitlán, Veracruz, Mexico
Ulises Juárez-Martínez Tecnológico Nacional de México/I. T. Orizaba, Orizaba,
Veracruz, Mexico
Angel Lagares-Lemos Department of Computer Science, Universidad Carlos III
de Madrid, Madrid, Spain
Miguel Lagares-Lemos Department of Computer Science, Universidad Carlos III
Olga Lucia Larrea-Serna Departamento de Calidad y Producción, Facultad de
Ciencias Económicas y Administrativas, Instituto Tecnológico Metropolitano,
Medellín, Antioquia, Colombia
Carolina Lima-Rivera Escuela de Ingeniería Industrial, Universidad del Valle,
Cali, Colombia
Carlos Eric Llorente-Peralta Tecnológico Nacional de México, Instituto
Tecnológico de Tijuana, Tijuana, Mexico
M. Lucila Morales-Rodríguez Tecnológico Nacional de México, Instituto
Tecnológico de Ciudad Madero, Ciudad Madero, Tamaulipas, Mexico
Jenny Lorena Luna-Eraso Universidad de Nariño, Pasto, Colombia
Asdrúbal López-Chau Universidad Autónoma Del Estado de México, Centro
Universitario UAEM Zumpango, Estado de México, Mexico
Roberto López-Rendón Universidad Autónoma del Estado de México, Toluca,
Mexico
Julián López-Velásquez Facultad de Ingeniería, Departamento de Instrumentación
y Control, Politecnico Colombiano, Medellín, Antioquia, Colombia
Isaac Machorro-Cano Universidad del Papaloapan, Tuxtepec, Oaxaca, Mexico
Martín Mancilla-Gómez Faculty of Accounting and Administration, Universidad
Veracruzana Campus Ixtaczoquitlán, Ixtaczoquitlán, Veracruz, Mexico
Heidy Marisol Marin-Castro Cátedras CONACYT, Facultad de Ingeniería y
Ciencias, Universidad Autónoma de Tamaulipas, Victoria, Tamaulipas, México
Erasto Alfonso Marín-Lozano Universidad Veracruzana, Xalapa, Veracruz,
México
Jonathan Alfonso Mata-Torres Unidad Académica Multidisciplinaria Reynosa-
RODHE, Universidad Autónoma de Tamaulipas, Reynosa, Tamaulipas, México
Santiago Medina Hurtado Universidad Nacional de Colombia, Medellín,
Colombia
xxii Contributors
Jezreel Mejía-Miranda Centro de Investigación en Matemáticas CIMAT, A.C,

Guanajuato, Mexico
Liliana Adriana Mendoza-Saboyá ISLP, International Statistics Institute, Bogotá,
Colombia
Diana María Montoya-Quintero Departamento de Calidad y Producción,
Facultad de Ciencias Económicas y Administrativas, Instituto Tecnológico
Metropolitano, Medellín, Antioquia, Colombia
Adrián Salvador Morales-García Department of Industrial Engineering and
Manufacturing, Universidad Autónoma de Ciudad Juárez, Ciudad Juárez, Chihuahua,
Mexico
Jorge Mulia-Rodríguez Universidad Autónoma del Estado de México, Toluca,
Mexico
Juan Carlos Nuñez-Dorantes Tecnológico Nacional de México/I. T. Orizaba,
Orizaba, Mexico
Juan Carlos Osorio-Gómez Valle del Cauca, Escuela de Ingeniería Industrial,
Logistic and Production Research Group, Universidad del Valle, Cali, Colombia
Josué Pacheco-Ortiz Tecnológico Nacional de México/I. T. Orizaba, Orizaba,
Veracruz, Mexico
José Fernando Padrón-Tristán Tecnológico Nacional de México, Instituto
Tecnológico de Tijuana, Tijuana, Mexico
Mario Andrés Paredes-Valverde Instituto Tecnológico Superior de Teziutlán,
Teziutlán, Puebla, México
S. Gustavo Peláez-Camarena Tecnológico Nacional de México/ IT Orizaba,
Yuliana Perez-Gallardo Department of Computer Science, Universidad Carlos III
Ulises Manuel Ramirez-Alcocer Unidad Académica Multidisciplinaria Reynosa-
RODHE, Universidad Autónoma de Tamaulipas, Reynosa, Tamaulipas, México
Rocío Ramos-Hernández Tecnológico Nacional de México/I. T. Orizaba, Orizaba,
Mexico
Nelson Rangel-Valdez Cátedras CONACyT/Tecnológico Nacional de México,
Instituto Tecnológico de Ciudad Madero, Ciudad Madero, Tamaulipas, Mexico
Lisbeth Rodríguez-Mazahua Tecnológico Nacional de México/ IT Orizaba,
Nidia Rodríguez-Mazahua Tecnológico Nacional de México/ IT Orizaba,
Contributors xxiii
Adolfo Rodríguez-Parada Faculty of Accounting and Administration, Univer-

sidad Veracruzana Campus Ixtaczoquitlán, Ixtaczoquitlán, Veracruz, Mexico
José Luis Rodríguez-Álvarez Doctoral Program in Engineering Sciences, Instituto
Tecnológico y de Estudios Superiores de Occidente (ITESO), Tlaquepaque, Jalisco,
México
J. Rogushina Institute of Software Systems of National Academy of Sciences of
Ukraine, Kyiv, Ukraine
Daniela Sarria-Cruz Escuela de Ingeniería Industrial, Universidad del Valle, Cali,
Colombia
Conrado Augusto Serna-Úran Instituto Tecnológico Metropolitano, Medellín,
Antioquia, Colombia
Manuel Suárez-Gutiérrez Universidad Veracruzana, Xalapa, Veracruz, Mexico
José Luis Sánchez-Cervantes CONACYT—Tecnológico Nacional de México/I.
T. Orizaba, Orizaba, Veracruz, Mexico
Laura Nely Sánchez-Morales Tecnológico Nacional de México/I. T. Orizaba,
Orizaba, México
Cuauhtémoc Sánchez-Ramírez Tecnológico Nacional de México/I. T. Orizaba,
Orizaba, Mexico
Edgar Tello-Leal Facultad de Ingeniería y Ciencias, Universidad Autónoma de
Tamaulipas, Victoria, Tamaulipas, México
Francisco Villanueva-Mejía Instituto Tecnológico de Aguascalientes, Aguas-
calientes, Mexico
Ángel Rodrigo Vélez-Bedoya Fundación Universitaria CEIPA Business School,
Antioquia, Colombia
Julian Andres Zapata-Cortes Fundación Universitaria CEIPA, Antioquia,
Colombia
List of Figures
Fig. 1.1 Model of the process choreography within a collaboration

diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Fig. 1.2 Overview of the methodology approach . . . . . . . . . . . . . . . . . . . 10
Fig. 1.3 Example of a bag-of-words at the case level generated
by method 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Fig. 1.4 Intra-organizational business process model discovered
for the M-Repair organization . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
Fig. 1.5 Process choreography discovered from the merged event
log . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Fig. 2.1 Traditional CAT process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
Fig. 2.2 Integration of association rule mining in the item selection
phase of the CAT process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Fig. 2.3 Comparison of Apriori, FPGrowth, PredictiveApriori
and Tertius algorithms in terms of support for Exa1 data set . . . 46
Fig. 2.4 Comparison of Apriori FPGrowth, PredictiveApriori
and Tertius algorithms in terms of confidence for Exa1
data set . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
and Tertius algorithms in terms of time for Exa1 data set . . . . . 47
data set . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
data set . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
xxv
xxvi List of Figures

Fig. 3.1 Uncertain sets and their membership functions: (a) ξ1
and μ1 ; (b) ξ2 and μ2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
Fig. 3.2 Membership functions for the unsharp concepts most,
young, and tall are represented, respectively, through
λ(x), ν(y), and μ(z) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
Fig. 3.3 Histograms, from left to right, of the Equity Loss Ratio
and Annual profitability of the investment funds . . . . . . . . . . . . 64
Fig. 3.4 The diagram is read clockwise starting at Financial Data.
Liu’s uncertain data mining poses a linguistic summarizer
on dataset A to evaluate the performance of the investment
funds. Implementation of the summarizer is general
in scope to ease the fit toward other problems . . . . . . . . . . . . . . 67
Fig. 3.5 Equity Loss Ratio vs Annual Profitability on investment
funds. A posteriori human verification through filter 1
of the logical sense for the first linguistic summary found
by the summarizer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
Fig. 3.6 Equity Loss Ratio versus Annual Profitability
on investment funds. A posteriori human verification
through filters 1 and 2 of the logical sense for the first
linguistic summary found by the summarizer . . . . . . . . . . . . . . . 70
Fig. 3.7 Estimation of the truth values T and probability p
for the linguistic summaries found. The summaries
describe the performance of investment funds
in Colombia in the period of study. Calculations of T
are made under the Liu’s uncertainty framework,
and the summaries, written in Human language, are found
by implementing the linguistic summarizer in (3.18) . . . . . . . . . 72
Fig. 4.1 Measurements of distances by Time of Flight (TOF)
of the ultrasonic sensors. Source Authors . . . . . . . . . . . . . . . . . . 78
Fig. 4.2 Map-Bot architecture for mapping of indoor
environments. Source Authors . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
Fig. 4.3 Four-tier architecture of the mobile robot Walling. Source
Authors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
Fig. 4.4 Electronic diagram of the SCoI wireless communications
subsystem. Source Authors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
Fig. 4.5 Distribution of components on the PCB. Source Authors . . . . . 83
Fig. 4.6 Typical navigation path of the robotic agent Walling
in a structured environment. Source Authors . . . . . . . . . . . . . . . 84
Fig. 4.7 Electronic diagram for the exteroceptive perception
and wireless communications of the Walling robot.
Source Authors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
Fig. 4.8 Walling robot fuzzy navigation controller block diagram.
Source Authors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
List of Figures xxvii
Fig. 4.9 Sketch of the Walling robot control structure. Source

Authors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
Fig. 4.10 Navigation tests on the Walling robot, a path of 560 cm
with a spurious reading of zero centimeters from the S0
sensor after traveling 400 cm. Source Authors . . . . . . . . . . . . . . 87
Fig. 4.11 Navigation tests on the Walling robot, a path of 560 cm
with a false reading greater than 100 cm after 420 cm
of travel. Source Authors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
Fig. 4.12 Specular reflection-free navigation in the Walling robot.
Source Authors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
Fig. 4.13 Partition of linguistic variables in fuzzy sets and control
surface of the fuzzy navigation system of the Walling
robot. Source Authors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
Fig. 4.14 Visual environment and “Adquisicion” mode of operation
of the Vespucci mapping module. Source Authors . . . . . . . . . . . 93
Fig. 5.1 Publication trend in the area of systems dynamic (a),
system dynamic supply chain (b), and c systems dynamic
food supply chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
Fig. 5.2 Visualization of network clusters of research
topics in publications related to system dynamics
with application in the food supply chain, period 2007–
2020. Note The minimum number of occurrences
of a keyword is two . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
Fig. 5.3 Countries with reported research on system dynamic
in the food supply chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
Fig. 5.4 Methodological approach (Aracil 1995) . . . . . . . . . . . . . . . . . . . 105
Fig. 5.5 Causal diagram of beekeeping production . . . . . . . . . . . . . . . . . 106
Fig. 5.6 Reinforcement loop R1 and R2 . . . . . . . . . . . . . . . . . . . . . . . . . . 106
Fig. 5.7 Balancing loop B1 y B2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
Fig. 5.8 Balancing loop B3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
Fig. 5.9 Forrester diagram of the beekeeping production
in Vichada. a Bees for breeding and Bees for honey,
bHoney hive inventory, Virgin honey collection center,
Pasteurized honey, Diversified product . . . . . . . . . . . . . . . . . . . . 108
Fig. 5.10 Behavior at the stock of bees for breeding in two scenarios . . . 111
Fig. 5.11 Behavior at the stock of bees for honey in two scenarios . . . . . . 111
Fig. 5.12 Behavior at the flow bee feedback in two scenarios . . . . . . . . . . 112
Fig. 5.13 Behavior at the flow sales of bees in the two scenarios . . . . . . . 112
Fig. 5.14 Behavior in the wax inventory in the two scenarios . . . . . . . . . . 113
Fig. 5.15 Inventory behavior of pasteurized honey, collection center
honey, and diversified product inventory in scenario 1 (a)
and scenario 2 (b) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
Fig. 5.16 Virgin honey collection center inventory . . . . . . . . . . . . . . . . . . . 115
Fig. 5.17 Pasteurized honey inventory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
Fig. 5.18 Inventory diversified product inventory . . . . . . . . . . . . . . . . . . . . 115
xxviii List of Figures
Fig. 6.1 Proposed model and relationships among variables . . . . . . . . . . 126

Fig. 6.2 Evaluated model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
Fig. 7.1 System architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146
Fig. 7.2 Energy flow of transformation centers . . . . . . . . . . . . . . . . . . . . . 147
Fig. 7.3 Types of reports and data dependencies . . . . . . . . . . . . . . . . . . . 148
Fig. 7.4 Analysis of energy balances . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
Fig. 7.5 Side effect of discarding measurements with quality bit
other than ‘00’ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
Fig. 7.6 Apparent profits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
Fig. 7.7 Data from TC1 to TC4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
Fig. 7.8 Data from TC5 to TC7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
Fig. 7.9 Reading rates of September . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152
Fig. 7.10 Reading rate of TC1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
Fig. 7.11 Reading rate of TC3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
Fig. 7.12 UTLB readings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
Fig. 7.13 Availability of reports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
Fig. 7.14 UTLA analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
Fig. 7.15 UTLB analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
Fig. 7.16 IoT reference architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
Fig. 7.17 Main concepts of the SSN ontology . . . . . . . . . . . . . . . . . . . . . . 160
Fig. 7.18 INDIGO software architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
Fig. 7.19 Major parts of SSN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
Fig. 7.20 Software architecture layers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
Fig. 7.21 Indigo home . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
Fig. 7.22 Indigo current data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
Fig. 7.23 Indigo read rate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164
Fig. 7.24 Indigo multiple dashboards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164
Fig. 7.25 Indigo datamodel menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
Fig. 8.1 Model for knowledge acquisition from Twitter . . . . . . . . . . . . . . 173
Fig. 8.2 Methodology model applied . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174
Fig. 8.3 The cluster of keywords . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174
Fig. 8.4 Server cluster configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178
Fig. 8.5 Summary of the pre-processing data layer . . . . . . . . . . . . . . . . . 179
Fig. 8.6 Tweets captured between June 11 and 13 of 2019 . . . . . . . . . . . 180
Fig. 8.7 Kibana Dashboard . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
Fig. 8.8 Veracity degree . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184
Fig. 8.9 Word Cloud for most used hashtags . . . . . . . . . . . . . . . . . . . . . . 187
Fig. 8.10 Hashtag correlation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188
Fig. 9.1 Layer model for the conservation of tacit knowledge
for DEA methodology. Source The authors . . . . . . . . . . . . . . . . 198
Fig. 9.2 Layer of human knowledge. Source The authors . . . . . . . . . . . . 206
Fig. 9.3 Layer of elements to find DEA processes. Source The
authors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208
Fig. 9.4 Layer of declarative knowledge elements. Source The
authors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208
List of Figures xxix
Fig. 9.5 Layer of cognitive knowledge. Source The authors . . . . . . . . . . 209

Fig. 9.6 Layer of elements of knowledge transformation
into processing. Source The authors . . . . . . . . . . . . . . . . . . . . . . 210
Fig. 9.7 Technology transfer of tacit knowledge in the DEA.
Source The authors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
Fig. 10.1 Proposed architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221
Fig. 10.2 Generation method for historical data with max entries (s
stands for satisfaction) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227
Fig. 11.1 Inference system based on compensatory fuzzy logic
(Espín-Andrade et al. 2014) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243
Fig. 11.2 Restrictions and interpretation criteria (Alonso et al. 2015) . . . 245
Fig. 11.3 Workflow for the order-picking problem using
Eureka-Universe (Padrón-Tristán et al. 2020) . . . . . . . . . . . . . . . 248
Fig. 11.4 The balance between precision and interpretability
(Cpałka 2017) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253
Fig. 11.5 Improving the balance between precision
and interpretability (Megherbi et al. 2019) . . . . . . . . . . . . . . . . . 254
Fig. 11.6 Approaches to design rules considering accuracy
and interpretability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255
Fig. 11.7 Execution of VS for the BPP (Padrón-Tristán 2020) . . . . . . . . . 257
Fig. 11.8 Results of the search in the number of papers using
eight different queries combining four keywords
(interpretability, interpretable, accuracy, and fuzzy)
and two filters (review type and keyword location) . . . . . . . . . . 258
Fig. 11.9 Results of the search in several papers (blue) and citations
(red) per year, using two queries that search in the paper
title and differing in the keyword fuzzy . . . . . . . . . . . . . . . . . . . . 258
Fig. 11.10 Number of research papers per journal . . . . . . . . . . . . . . . . . . . . 259
Fig. 12.1 Proposed model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273
Fig. 12.2 Evaluated model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 280
Fig. 13.1 Operational risk management system in supply chains
(Manotas et al. 2014) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291
Fig. 13.2 Methodological design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291
Fig. 13.3 Risk identification approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292
Fig. 13.4 Probability and impact matrix (Osorio-Gomez et al. 2018) . . . . 293
Fig. 13.5 Methodological approach to risk prioritization
(Osorio-Gomez et al. 2018) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294
Fig. 13.6 Reverse logistics network to lead-acid batteries . . . . . . . . . . . . . 295
Fig. 13.7 The battery recovery process in a Colombian company . . . . . . . 296
Fig. 13.8 Battery recovery process (https://www.ambientebogota.
gov.co/, 2008) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 296
Fig. 13.9 Risk probability-impact matrix . . . . . . . . . . . . . . . . . . . . . . . . . . 298
Fig. 13.10 Cause effect diagram to operational risk in reverse
logistics of lead-acid batteries . . . . . . . . . . . . . . . . . . . . . . . . . . . 300
xxx List of Figures
Fig. 13.11 Questionnaire for knowledge validation in new income

and monitoring in old workers . . . . . . . . . . . . . . . . . . . . . . . . . . . 302
Fig. 13.12 Personal protective equipment . . . . . . . . . . . . . . . . . . . . . . . . . . . 303
Fig. 14.1 Most common industrial applications of sugarcane
vinasses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 312
Fig. 14.2 A conceptual model of the vinasse-based LF supply chain . . . . 313
Fig. 14.3 Illustration of the vinasse-based LF supply chain . . . . . . . . . . . . 315
Fig. 14.4 Different ways to system feedback . . . . . . . . . . . . . . . . . . . . . . . 315
Fig. 14.5 Mass balance of LF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 316
Fig. 14.9 Behavior of utilities within the vinasse-based LF supply
chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 320
Fig. 14.10 Causal diagram of vinasse-based LF production . . . . . . . . . . . . 321
Fig. 14.11 Feedback loop used in outlier test . . . . . . . . . . . . . . . . . . . . . . . . 324
Fig. 14.12 Model validation results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325
Fig. 14.13 Molasses storage at XYZ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 326
Fig. 14.14 Daily ethanol production at XYZ . . . . . . . . . . . . . . . . . . . . . . . . . 327
Fig. 14.15 Daily ethanol production versus vinasse generated . . . . . . . . . . 328
Fig. 14.16 Vinasse inventory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 329
Fig. 14.17 Inventory for 500 h . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 329
Fig. 14.18 Behavior of LF production on a yearly basis . . . . . . . . . . . . . . . 330
Fig. 14.19 LF inventory behavior under different LF demand
scenarios . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 332
Fig. 15.1 Data set with 24 queries and 2 fragments . . . . . . . . . . . . . . . . . . 346
Fig. 15.2 Algorithm 1. Generation of data sets . . . . . . . . . . . . . . . . . . . . . . 346
Fig. 15.3 Algorithm 2. Get PPM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 347
Fig. 15.4 Results of the Recall metric for 24 queries data sets . . . . . . . . . 349
Fig. 15.5 Results of the Precision metric for 24 queries data sets . . . . . . . 350
Fig. 15.6 Results of the ROC Area metric for 24 queries dataset . . . . . . . 350
Fig. 15.7 Results of the F-Measure metric for 24 queries dataset . . . . . . . 351
Fig. 15.8 Decision tree created by J48 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 352
Fig. 15.9 Results of decision tree algorithms for 24 queries and 2
fragments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 353
Fig. 15.10 Results of decision trees algorithms for 24 queries and 3
fragments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 354
fragments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 354
fragments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 355
fragments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 355
fragments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 356
List of Figures xxxi

fragments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 357
fragments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 358
Fig. 15.17 Horizontal Fragmentation method diagram . . . . . . . . . . . . . . . . . 359
Fig. 16.1 Increase in loan of P2P lending since January 2016
(Superintendencia Financiera de Colombia 2019) . . . . . . . . . . . 366
Fig. 16.2 Personal loans default versus P2P lending default
(Superintendencia Financiera de Colombia 2019) . . . . . . . . . . . 367
Fig. 16.3 Accuracy versus depth on decision trees . . . . . . . . . . . . . . . . . . . 380
Fig. 16.4 Sub-tree from the optimized decision tree . . . . . . . . . . . . . . . . . . 380
Fig. 17.1 Generalized algorithm of task thesaurus generation . . . . . . . . . . 393
Fig. 17.2 Use of OntoSearch for selection of ontological classes
to initial task thesaurus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 396
Fig. 17.3 Expansion of thesaurus on base of taxonomic relation . . . . . . . . 397
Fig. 17.4 Expansion of thesaurus on base of various types
of hierarchical relation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 397
Fig. 17.5 Use of task thesaurus for semantic retrieval in MAIPS . . . . . . . 400
Fig. 17.6 e-VUE ontology and user interface . . . . . . . . . . . . . . . . . . . . . . . 401
Fig. 18.1 Interactions in the atom-atom model between different
molecules where the atom a = 1 of molecule i interacts
with the atoms of molecule j and so on with the other atoms . . 410
Fig. 18.2 Periodic boundary conditions in a periodic
two-dimensional system The shaded cell corresponds
to the central cell . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 412
Fig. 18.3 Lennard-Jones potential. The blue line represents
the attraction of atoms and the red line their repulsion . . . . . . . . 414
Fig. 18.4 NVT algorithm incorporating NHC thermostats. The
NVE ensemble is located in the central part . . . . . . . . . . . . . . . . 415
Fig. 18.5 Algorithm of the application of the Nosé-Hoover chain
thermostat . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 416
Fig. 18.6 Processing diagram of a simulation with molecular
dynamics and big data analytics . . . . . . . . . . . . . . . . . . . . . . . . . . 417
Fig. 18.7 Big data logical diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 419
Fig. 18.8 Big data physical diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 420
Fig. 18.9 CQL sentences used in the creation of the database
schema for Cassandra™ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 421
Fig. 18.10 Initial configuration of a 2048 particle LJ fluid system
with temperature 2.0 and density 0.7 . . . . . . . . . . . . . . . . . . . . . . 423
Fig. 18.11 Representation of: A Kinetic energy; B Potential energy
and, C Temperature calculated to balance the system.
Note the initial values of each of them and later
the constant values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 423
Fig. 18.12 3D model of the new positions and velocities . . . . . . . . . . . . . . . 424
xxxii List of Figures
Fig. 18.13 Representation of: A Kinetic energy; B Potential

energy and, C Calculated temperature in a production
dynamic with 40,000 steps. In each one of them,
the constant conservation of the calculated values is
noted, characteristic of the NVT ensemble . . . . . . . . . . . . . . . . . 424
Fig. 18.15 Initial configuration of a 10,976 particle LJ fluid system
Fig. 18.20 Initial configuration of a 23,328 particle LJ fluid system
Fig. 19.1 Family entrepreneurship . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 443
Fig. 19.2 Financial need . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 443
Fig. 19.3 Economic solvency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 444
Fig. 19.4 Variance contribution of each component . . . . . . . . . . . . . . . . . . 445
Fig. 19.5 Panel A origin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 446
Fig. 19.6 Panel B origin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 447
Fig. 19.7 Panel C origin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 447
Fig. 19.8 Panel D origin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 448
Fig. 19.9 Panel E origin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 448
Fig. 19.10 Panel F origin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 449
Fig. 19.11 Panel A generational change . . . . . . . . . . . . . . . . . . . . . . . . . . . . 449
Fig. 19.12 Panel B generational change . . . . . . . . . . . . . . . . . . . . . . . . . . . . 450
Fig. 19.13 Panel C generational change . . . . . . . . . . . . . . . . . . . . . . . . . . . . 450
List of Figures xxxiii
Fig. 19.14 Panel D generational change . . . . . . . . . . . . . . . . . . . . . . . . . . . . 451

Fig. 19.15 Panel E generational change . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 451
Fig. 19.16 Panel F generational change . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 452
Fig. 19.17 Tree for economic solvency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 453
Fig. 19.18 Generation of founders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 454
Fig. 19.19 Second generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 455
Fig. 19.20 Third generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 455
Fig. 20.1 Collaborative scheme for the VMI . . . . . . . . . . . . . . . . . . . . . . . . 466
Fig. 20.2 Chromosome used in the genetic algorithm . . . . . . . . . . . . . . . . 467
Fig. 20.3 Crossover operator used in the genetic algorithm . . . . . . . . . . . . 467
Fig. 20.4 Mutation operator used in the genetic algorithm . . . . . . . . . . . . 474
Fig. 20.5 Best individual obtained by the genetic algorithm . . . . . . . . . . . 476
Fig. 20.6 Solution representation of using the VRP with time
windows . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 478
Fig. 21.1 Educational resource R298—Perimeter and area
of geometric figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 501
Fig. 21.2 Educational resource R280—Area formulas for geometric
shapes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 503
List of Tables
Table 1.1 Example of a case in the M-Parts participant’s event log . . . . 14

Table 1.2 Example of a case in the M-Repair participant’s event log . . . 15
Table 1.3 Score matrix at the case level . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Table 1.4 Score matrix at the activity level . . . . . . . . . . . . . . . . . . . . . . . . 17
Table 1.5 Type and sub-type of discovered tasks for the event log
BPMNI of the M-Repair participant . . . . . . . . . . . . . . . . . . . . . 17
Table 1.6 Type and sub-type of discovered tasks for the event log
BPMNR of the M-Parts participant . . . . . . . . . . . . . . . . . . . . . . 18
Table 2.1 Related works (A) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
Table 2.2 Related works (B) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
Table 2.3 Related works (C) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
Table 2.4 Test results for Apriori and FPGrowth for the Exa1 data
set . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
set . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
set . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
Table 2.7 Test results for PredictiveApriori and Tertius
for the Exa1 data set . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
Table 2.10 Test results for Apriori and FPGrowth in terms of Lift
for the Exa1, Exa2 and Exa3 data set . . . . . . . . . . . . . . . . . . . . 45
Table 3.1 Linguistics and m.f. of the uncertain quantifier . . . . . . . . . . . . 64
Table 3.2 Linguistics and m.f. of the uncertain subject . . . . . . . . . . . . . . 65
Table 3.3 Linguistics and m.f. of the uncertain predicate . . . . . . . . . . . . . 65
Table 3.4 Results: linguistic summaries—part I (Colombian
investment funds) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
Table 3.5 Results: linguistic summaries—part II (Colombian
investment funds) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
xxxv
xxxvi List of Tables
Table 3.6 Linguistic summaries - Truth value T vs. probability

measure p . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
Table 4.1 Technical specifications of the SFR02 sensor. Source
(Acosta 2010) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
Table 4.2 Characterization of linguistic variables for the fuzzy
navigation controller of the robot walling . . . . . . . . . . . . . . . . . 90
Table 4.3 Simplified FAM for the fuzzy navigation controller
of the robot walling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
Table 5.1 State of the art on the systems dynamic in food supply
chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
Table 5.2 Description and role of major variables used to model
the Beekeeping Supply Chain . . . . . . . . . . . . . . . . . . . . . . . . . . 109
Table 6.1 Questionnaire validation indexes . . . . . . . . . . . . . . . . . . . . . . . . 128
Table 6.2 Model fit and quality indexes . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
Table 6.3 Gender versus years of experience . . . . . . . . . . . . . . . . . . . . . . 132
Table 6.4 Number of employees versus job position . . . . . . . . . . . . . . . . 132
Table 6.5 Validation of latent variables . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
Table 6.6 Descriptive analysis of the items . . . . . . . . . . . . . . . . . . . . . . . . 134
Table 6.7 Direct effects contribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
Table 6.8 Total effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
Table 6.9 Sensitivity analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
Table 8.1 Selection of related works about violence in social media . . . 172
Table 8.2 List of user key Twitter accounts . . . . . . . . . . . . . . . . . . . . . . . . 176
Table 8.3 Top 20 keyword validation list . . . . . . . . . . . . . . . . . . . . . . . . . . 177
Table 8.4 Detail of node configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . 178
Table 8.5 Principal emoticons and special characters deleted . . . . . . . . . 182
Table 8.6 Descriptive analysis of frequencies at the federal states’
level . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
Table 8.7 Descriptive analysis of frequencies at the metropolitan
zones level . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
Table 8.8 Classification range of analysis intervals at a state level . . . . . 186
Table 8.9 Classification of analysis intervals at metropolitan zones . . . . 186
Table 8.10 Intervals for the indicators created at the federal entity
level . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
Table 8.11 Intervals for the indicators created at metropolitan zones . . . . 191
Table 9.1 Multiplicative and envelopment form of the models . . . . . . . . 212
Table 10.1 Instance composition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225
Table 10.2 Range value considered for the sensors . . . . . . . . . . . . . . . . . . 225
Table 10.3 Training set of the historical data, HE . . . . . . . . . . . . . . . . . . . . 226
Table 10.4 Test set of the historical data, HP . . . . . . . . . . . . . . . . . . . . . . . 226
Table 10.5 Initial configuration of weights and thresholds used
by the PDA strategy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228
Table 10.6 ELECTRE III parameters’ values obtained through
PDA strategy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228
Table 10.7 Chosen reference value for outranking . . . . . . . . . . . . . . . . . . . 230
List of Tables xxxvii
Table 10.8 CFL rules generated using EUREKA UNIVERSE . . . . . . . . . 231

Table 10.9 Value of S1 for HE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232
Table 10.10 Value of S2 for HP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232
Table 10.11 Description of the success case . . . . . . . . . . . . . . . . . . . . . . . . . 232
Table 10.12 Memory consumption in comparison with other
mechanisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233
Table 11.1 Comparison of surveys and reviews related to this work . . . . . 246
Table 11.2 MF parameters values of linguistic states for premise 1 . . . . . 250
Table 11.3 Sample of classification accuracy for premise 1 . . . . . . . . . . . . 250
Table 11.4 Accuracy of premises obtained . . . . . . . . . . . . . . . . . . . . . . . . . 250
Table 11.5 Skill comparison by tasks with the principal fuzzy ways
to use productively the natural language Logic . . . . . . . . . . . . 252
Table 11.6 Comparison of selected papers on accuracy
and interpretability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 260
Table 12.1 Quality planning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 270
Table 12.2 Quality management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 270
Table 12.3 JIT benefits for human resources . . . . . . . . . . . . . . . . . . . . . . . . 271
Table 12.4 JIT Economic benefits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272
Table 12.5 Descriptive analysis of the sample . . . . . . . . . . . . . . . . . . . . . . 277
Table 12.6 Industrial sectors and number of employees . . . . . . . . . . . . . . . 277
Table 12.7 Latent variable coefficients . . . . . . . . . . . . . . . . . . . . . . . . . . . . 278
Table 12.8 Descriptive analysis of items and variables . . . . . . . . . . . . . . . 279
Table 12.9 Sum of indirect effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282
Table 12.10 Total effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283
Table 13.1 Linguistic scale for the risk identification and fuzzy
equivalence for FQFD (Pastrana-jaramillo
and Osorio-gómez 2018) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292
Table 13.2 Validation and weighted averages of operational risks . . . . . . 297
Table 13.3 Internal variables and their relative importance . . . . . . . . . . . . 299
Table 13.4 Weight of the how’s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299
Table 13.5 Results of prioritization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299
Table 14.1 Physicochemical characteristics of raw vinasse . . . . . . . . . . . . 316
Table 14.2 Variables involved in the vinasse-based LF supply chain . . . . 318
Table 14.3 Scenarios for sensitivity analysis of LF demand . . . . . . . . . . . 331
Table 15.1 Comparative table of works on horizontal
fragmentation (A) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 344
Table 15.2 Comparative table of works on horizontal
fragmentation (B) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 345
Table 15.3 Results of decision trees algorithms with 50 queries
for two fragments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 351
Table 15.4 Results of decision tree algorithms with 50 queries
for three fragments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 351
Table 15.5 Results of decision tree algorithms with 50 queries
for four fragments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 352
xxxviii List of Tables
Table 15.6 Results of decision trees algorithms with 50 queries

for five fragments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 352
Table 15.7 Results of decision tree algorithms for 24 and 50 queries . . . . 358
Table 16.1 Health category for microcredits, according default age . . . . . 372
Table 16.2 Transition matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 373
Table 16.3 Date variables conversion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 376
Table 16.4 Variables with null data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 377
Table 16.5 Variables variance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 378
Table 16.6 Evaluation metrics comparison of the five techniques . . . . . . . 379
Table 16.7 Optimized decision tree performance . . . . . . . . . . . . . . . . . . . . 380
Table 16.8 Variables and its importance in prediction . . . . . . . . . . . . . . . . 381
Table 18.1 Comparative table of thermostats and architecture
in which it is designed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 409
Table 19.1 Linear decomposition based on the original variables . . . . . . . 445
Table 19.2 Main components composition . . . . . . . . . . . . . . . . . . . . . . . . . 445
Table 19.3 Levels of the variables in the solvency classification . . . . . . . . 453
Table 20.1 Summary of selected works about IRPTW . . . . . . . . . . . . . . . . 469
Table 20.2 Data used to test the model . . . . . . . . . . . . . . . . . . . . . . . . . . . . 475
Table 20.3 Inventory in each customer for every period . . . . . . . . . . . . . . 477
Table 20.4 Cost components of the IRPTW solution . . . . . . . . . . . . . . . . . 478
Table 20.5 Comparison between IRPTW and VRPTW . . . . . . . . . . . . . . . 478
Table 21.1 Comparative analysis of related works . . . . . . . . . . . . . . . . . . . 488
Table 21.2 APIs for emotion detection, sentiment analysis,
and named-entities recognition . . . . . . . . . . . . . . . . . . . . . . . . . 496
Table 21.3 Programming languages for emotion detection,
sentiment analysis, and named-entities recognition . . . . . . . . . 499
Table 21.4 Results of the comparative analysis among APIs
for scenario 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 502
Table 21.5 Results of the comparative analysis of APIs in scenario 2 . . . 504
Part I
Industrial Applications
Chapter 1
Merging Event Logs
for Inter-organizational Process Mining
Jaciel David Hernandez-Resendiz, Edgar Tello-Leal,

Heidy Marisol Marin-Castro, Ulises Manuel Ramirez-Alcocer,
and Jonathan Alfonso Mata-Torres
Abstract In an inter-organizational environment, the discovery of process choreog-

raphy is challenging because the different organizations involved have to put together
their partial knowledge about the overall collaborative business process. This chapter
presents a methodology to merge the event logs of the different partners of a collab-
orative business process, in order to serve as input for the process mining algorithm.
On the one hand, the methodology consists of a method set for searching the corre-
lation between events of the log of different partners involved in the collaboration.
These methods are implemented at the trace level and the activity level. On the
other hand, the methodology consists of a set of methods and rules for discovering
process choreography. From the knowledge gained by the above methods, message-
type tasks are identified and marked in each event log, then using a formal set of
rules, the message task sub-type (send or receive) is discovered. Finally, links using
message sequence flow connectors between message tasks identified as pair activi-
ties in event logs are automatically defined. The proposed approach is tested using
a real-life event log that confirms their effectiveness and efficiency in the automatic
J. D. Hernandez-Resendiz · U. M. Ramirez-Alcocer · J. A. Mata-Torres

Unidad Académica Multidisciplinaria Reynosa-RODHE, Universidad Autónoma de Tamaulipas,
Reynosa, Tamaulipas, México
e-mail: a2183728004@alumnos.uat.edu.mx
U. M. Ramirez-Alcocer
J. A. Mata-Torres
E. Tello-Leal (B)
Facultad de Ingeniería y Ciencias, Universidad Autónoma de Tamaulipas, Victoria, Tamaulipas,
México
e-mail: etello@docentes.uat.edu.mx
H. M. Marin-Castro
Cátedras CONACYT, Facultad de Ingeniería y Ciencias, Universidad Autónoma de Tamaulipas,
Victoria, Tamaulipas, México
e-mail: hmarin@conacyt.mx
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 3

J. A. Zapata-Cortes et al. (eds.), New Perspectives on Enterprise Decision-Making
Applying Artificial Intelligence Techniques, Studies in Computational Intelligence 966,
https://doi.org/10.1007/978-3-030-71115-3_1
4 J. D. Hernandez-Resendiz et al.
specification of message flows of the process choreography discovered, allowing to

build a collaborative business process model.
1.1 Introduction
Public and private organizations require proper knowledge asset management to

maintain a competitive advantage in current global markets. All-business are facing
global competition, and it is imperative that they reduce costs, improve their opera-
tions, and the relationships with customers, suppliers, and partners, trying to reduce
also delivering times through optimizing their logistics processes (Sousa et al. 2017).
The adoption of new technologies in this context encourages organizations to estab-
lish close relationships of integration, cooperation, and collaboration between them,
giving rise to inter-organizational collaborations (Pradabwong et al. 2015; Long
2017). This collaboration contributes to improving the efficiency of supply chain
management, which focuses on the inter-organizational management of goods flows
between independent companies in a supply chain (Sousa et al. 2017).
Supply chain collaboration enables chain members to take advantage of business
opportunities and enhance their competitiveness. This means that, in a supply chain
collaboration, two or more chain members working together to create a compet-
itive advantage by sharing information, joint decision making, and sharing bene-
fits of increased profitability that result from satisfying customer needs (Salam
2017; Simatupang and Sridharan 2018), as well as executing the collaborative
business processes (also called inter-organizational business processes) through
process-aware information systems (Dumas et al. 2018a).
In this sense, Business Process Management (BPM) is concerned with the inter-
actions between processes and information systems, standing out the modeling and
analysis of processes as important elements within of BPM approach (van der Aalst
2018). Thus, it can be considered that BPM is a set of methods, techniques, and tools
to identify, discover, analyze, redesign, execute, and monitor business processes
in order to optimize their performance (van der Aalst 2013; Dumas et al. 2018b).
Business processes are the core of BPM approaches, and various languages have
been introduced for business process modeling, with the Business Process Model
and Notation (BPMN) language (BPMN 2.0 2011), being the de facto standard for
business process notation.
In the BPM domain, inter-organizational collaboration implies a process-oriented
integration between heterogeneous and autonomous organizations, which can be
achieved through the definition and execution of collaborative business processes.
A collaborative business process defines the behavior of interactions between orga-
nizations and their roles from a global viewpoint (Tello-Leal et al. 2016; Barcelona
et al. 2018), that is, how they coordinate their actions and exchange information,
in order to make decisions together to achieve a common business goal (Tello-Leal
et al. 2016; Köpke et al. 2019). In inter-organizational collaboration, the interaction
between participants is done via message exchanges; the interaction is only archived
1 Merging Event Logs for Inter-organizational Process Mining 5
Fig. 1.1 Model of the

process choreography within
a collaboration diagram
by sending and receiving messages. In a BPMN-based collaboration diagram, the

pools represent specific process participants or roles, such as the supplier role or
customer role. The interactions of a set of business processes of multiple organiza-
tions are specified in a process choreography (Weske 2019). A collaboration diagram
describes the control flow layer of the choreography and depicts what messages and in
which sequence are exchanged. Figure 1.1 shows an interaction between a customer
and a supplier with respect to a request for a demand forecast. The customer sends
the forecast request to a chosen supplier, which internally processes it and sends
the forecast-request response, which then is evaluated internally by the customer.
Hence, choreographies have a central role in ensuring interoperability between
business processes (Weske 2019), each of which is performed by a participant in
inter-organizational collaboration.
One of the current challenges in inter-organizational collaborations is the analysis
of the execution of collaborative business processes. Process mining makes use of
recorded historical data (event log) of the execution of the business process instances
to discover and analyze as-is business process models (van der Aalst 2016). However,
current process mining approaches focus on analyzing the execution of the business
process from an intra-organizational perspective (Kalenkova et al. 2017; Nguyen
et al. 2019; Mehdiyev et al. 2020). An analysis of the execution of a collaborative
business process has high complexity, due to the correlation and synchrony that must
be identified in the interactions between the parties involved in the collaboration, as
well as to the difficulty to capture the behavior of artifacts (business documents, input,
and output objects) that are distributed among the collaborative processes. Moreover,
in an inter-organizational environment, the event logs are distributed over different
sources, that is; one event log per enterprise that participates in the collaboration.
Each event log encompasses partial information about the overall business process.
Therefore, inter-organizational process mining requires these historical data to be

merged into one structured event log.
In this book chapter, we propose an inter-organizational process mining for
discovering the interaction between business processes of multiple organizations via
message exchange. In particular, we propose a methodology that guides the proce-
dure of merging event logs and discovering process choreography. In this regard, we
propose the integration of event logs through a merge at the case level and the activity
level. This merge is performed by computing the cosine similarity at the case and
activity level, allowing the correlation of the message tasks to be discovered. Further-
more, we define a formal rule set to identify the message-type tasks in each event log
involved in the collaboration, as well as the sub-type of the message tasks (send or
receive), which enables us to discover the sense of the message flow. Finally, message
flow connectors are specified between message-type tasks contained in a diagram of
the collaborative business process, enabling the discovery of process choreography
among participants in inter-organizational collaboration. The proposed approach is
validated in a real scenario using an event log generated from the execution of a
purchase order management process.
The remainder of this chapter is organized as follows: Sect. 1.2 reports the results
achieved by implementing the proposed approach. Section 1.3 introduces the prelim-
inary concepts and definitions that support the proposed approach. Section 1.4 details
the phases that compose our methodology. Section 1.5 discusses the related work
found in the literature. Finally, the conclusion and future work are given in Sect. 1.6.
1.2 Related Work
In this section, we delineate the current literature from pre-processing and merging of
logs, and end-to-end flow analysis of related processes by correlation methods. The
technique proposed in Claes and Poels (2014) consists of an algorithm that searches
for links between the data of the different partners suggesting the rules to the user on
how to merge the data, and a method for configuring and executing the merge; as well
as their implementation in the process mining tool ProM (Van Dongen et al. 2005).
The algorithm discovers links between the two event logs to indicate which data in
both event logs are considered to belong to the same process instance. The merging
rules are formulated in terms of relations between attribute values at an event or
trace level in the two event logs. The rules are based on four relationship operators
and two logical operators (and/or). The approach is tested through seven scenarios
involving two artificially created and three real-life event logs, for demonstration
and evaluation of the method and algorithm. The test results show positive results
for effectiveness and efficiency.
In Raichelson et al. (2017), they describe an automated technique to merge event
logs supporting two granularity levels. They generate a merged log with a focus
on the case view and a merged log that reveals the end-to-end instance view. The
matching of cases is based on temporal relations and text similarity, both structured
and unstructured attributes. Similarly, (Bala et al. 2018) propose a semi-automatic

technique for discovering event relations that are semantically relevant for business
process monitoring. In this proposal, the challenge that is faced that the events are
available in different levels of granularity and that more than one event can correspond
to an activity, proposing identifiers for events and relationships that are relevant
to monitoring the process. The preceding, under the assumption that these events
contain data that are relevant to monitoring, but without prior knowledge of the
event schema, using a set of heterogeneous events as input. This approach can be
considered a contribution of a technique for the pre-processing of event logs for
subsequent analysis using process mining methods.
On the other hand, in Engel et al. (2016), the authors propose an approach to
associate Electronic Data Interchange (EDI) messages from different data sources
belonging to the same case based on correlation conditions. This correlation is based
on conjunctions and/or dis-junctions of attribute values. Such conditions are formu-
lated in reference to the attribute values of the messages. In this sense, (Pourmirza
et al. 2017) tackle the correlation problem by extracting models from logs without
relying on case identifiers or when events are not associated with a case identifier
by process graph rules, enabling detect which events belong to the same case. This
approach only relies on event names and timestamps in the log, no additional event
attributes are required.
Similarly, in Cheng et al. (2017) presents a graph-based method that follows the
filtering-and-verification principle, which aims to efficient event correlation analytics
over large logs on distributed platforms. This approach incorporates light-weight
filter units into candidate correlation rules to prune large numbers of non-interesting
rules before verification. In the verification phase, the correlation rules are modelled
as a graph and introduce a graph partitioning approach to decompose the potentially
correlated events into chunks by exploring efficient data locality assignment.
In Xu et al. (2018) presents an algorithm based on artificial immune algorithms
and simulated annealing algorithms to merge event log files generated by different
systems. The set of factors used in the affinity function, occurrence frequency, and
temporal relation can express the characteristics of matching cases more accurately
than some factors. The proposed algorithm has been implemented as a plugin into
the ProM platform (Van Dongen et al. 2005).
In summary, the approach we present here has advantages. First, it enables the
merging of event logs based on event messages correlation through rule and pattern
set. This renders to discover message tasks and the sub-type of the message task.
Second, it provides the two abstraction levels of both case and activity. Third, the
resulting combination of abstraction levels enables mining that is targeted at discov-
ering process choreography among participants in inter-organizational collaboration.
However, our proposal has the capacity to discover simple relationships, that is, one-
to-one and one-to-many relationships, which can be considered a limitation in the
algorithm. Then, it is necessary to define a set of rules so that the methods have the
ability to identify complex many-to-one or many-to-many relationships, in the traces
contained in each participant’s event log.
1.3 Preliminaries and Definitions
The process mining area that is aimed to discover, verify, and improve real business
processes from events available in process-aware information systems (van der Aalst
2016), that is, aims at extracting process knowledge from event logs. An event log
consists of a set of traces and each trace is composed of a sequence of events produced
by the execution of one case. Each event captures relevant data about the execution
of a given activity in the business process. Therefore, through the application of
process mining, organizations can discover how the processes were executed, verify
if the defined business rules and practices were followed, as well as gain insights into
bottlenecks, resource utilization, and other performance-related aspects of processes
(Rovani et al. 2015).
The merge criterion for our proposal is based on using cosine similarity to derive a
measure of traces similarity and activities similarity between two event logs. Cosine
similarity analyzes the similarity by measuring the cosine of the angle between the
two vectors of an inner product space (Han and Kamber 2012). It is measured by
the cosine of the angle between two vectors and determines whether two vectors are
pointing in roughly the same direction. The cosine similarity measure (Manning and
Raghavan 2008) cosSim(j,q) can be computed as follows (Eq. 1.1):
t
(Wi j .Wiq )
Cos Sim( j, q) = i=1 t (1.1)
t
( i=1 Wi j )2 .( i=1 Wiq )2
t
where i=1 (Wi j .Wiq ) is the sum of the scalar product of the weights of the term i
of
the word vector j by the weight of the term i of the word vector q, this divided by
t t
( i=1 Wi j )2 .( i=1 Wiq )2 , which is the square root of the sum of the weights of
the term i of the word vector j squared by the sum of the weights of the term i of the
word vector q squared.
The proposed approach is based on the following definitions, which make it
possible to formalize the phases of the methodology, which enables the fusion of
event logs and the discovery of the correlation of the messages exchanged in a
collaboration.
Definition 1.1 (Event log) Given an event log L, which is compound of a set of
cases T, where T are all instances of the business process execution. Each case T i
is composed of a finite set of activities A, which are the set of tasks contained in the
business process. These tasks are described by a set of At attributes (for example,
activity name, execution date, the user who executed it, among others), which detail
the context of the execution of the business process activities.
Definition 1.2 (The bag-of-words of cases) Refers to a vector of words that represents
each case T i in the event L. This bag-of-words is generated from the unique values
of the At attributes of each Aj T i activity.
Definition 1.3 (The bag-of-words of activities) Refers to a vector of words that

represent each case Aj activity. This bag-of-words is generated from the unique
values of the At k attributes composite Aj .
Definition 1.4 (Inter-organizational relationship) Indicates that two or more organi-

zations share event information from the same execution of a collaborative business
process. This relationship is identified in the events contained in the traces of the
event L and L’, which correspond to each participating organization in the inter-
organizational collaboration. In our proposal, two levels of the inter-organizational
relationship are contemplated: a) the relationship at the case level occurs when in
two cases T L and T’ L’, their similarity (calculated using the cosine measure) is
located above a threshold Ut, b) the relationship at the activity level occurs when in
two activities A T and A’ T’, their similarity (cosine measure) is above a threshold
U a.
Definition 1.5 (BPMN process model) This statement is based on the definition
proposed by (Augusto et al. 2019), in which a BPMN process model is a component
graph M = (i, o, T, G, E m ), where i is the start event, o is the end event, T is a
non-empty set of tasks, G+ ∪ Gx ∪ G* is the union of the set of AND gateways (G+ ),
the set of XOR gateways (Gx ) and the set of OR gateways (G*), and E m ⊆ (T ∪ G ∪
{i}) × (T ∪ G ∪ {o}) is the set of edges. Further, given g G, g is a split gateway if it
has more than one outgoing edges, or join gateway if it has more than one incoming
edges.
1.4 Methodology
The proposed methodology for merging event logs and discovering process choreog-
raphy in an inter-organizational collaboration environment is composed of processing
of event logs, identifying the correlation between events, and collaboration discovery
phases. Figure 1.2 illustrates the procedure of the proposed methodology, where the
event records of the business processes involved in inter-organizational collaboration
are used as input to phase 1, and as an output from phase 3, the collaborative business
process model discovered.
1.4.1 Phase 1: Processing of Event Logs
1.4.1.1 Method 1: Construction of Bags-of-Words
This method consists of building bags-of-words per event log, which allows each
bags-of-words to be represented in a matrix. This procedure is performed by every
organization (one event log per organization) involved in the collaborative business
Fig. 1.2 Overview of the methodology approach
process. For each event log L and L’ a matrix of bags-of-words BW, and BW ’ is
generated, where one row (a trace in the event log) of the matrix is a bag-of-word
of T L and T ’ L’, respectively. The length of each bag-of-words is equal to the
number of attributes that describe the activities contained in a trace. The matrices
BW and BW ’ are constructed as follows:
• Let us consider any case T i L.
• A bag-of-word BW i is built, with all the unique values of the At k attributes of
each task Aj T i , based on Definition 1.2.
• In BW i , stop-words are removed.
• The remaining words of BW i compose the bag-of-words T i .
• This procedure is carried out independently for the event logs L and L’.
1.4.1.2 Method 2: Generation of the Scoring Matrix
In this method for each of the BW i bags-of-words that belong to the BW bag-of-
words, they are compared to all the bags-of-words that compose the BW ’. This is
performed using the cosine similarity according to Definition 1.4. The above allows
us to generate a scoring matrix with the similarity between the vectors of BW and
BW ’.
1.4.2 Phase 2: Identifying the Correlation Between Events
1.4.2.1 Method 3: Selection at the Case Level
The selection of the even cases consists of finding the cases T L and T ’ L’ with
the closer the cosine value to 1. For each T i L case, the case T i ’ L’ with the
greater cosine similarity in the scoring matrix is selected. This allows us to identify
that in cases T L and T ’ L’ there is a relationship at the case level, according to
Definition 1.4. This process is performed for each trace of the event log L.
1.4.2.2 Method 4: Selection at the Activity Level
From the cases selected by Method 3, the tasks or events (activities) that contain
any coincident data are identified, allowing a relationship between the activities of
the traces to be defined, calculating their cosine similarity, that is, similarity at the
activity level, for which the following procedure is implemented:
• The attribute or attributes that can provide information on the relationship between
organizations are defined, for example, resource, timestamps, activity name, input,
or output objects.
• For each activity Aj T i and Aj ’ Ti’ of each pair of cases T i L and T i ’
L’ (which were selected in the previous method because their cosine similarity
value exceeds the U t threshold), two vectors of activities A and A’ are obtained,
which represent the bags-of-words BWA and BWA’ according to Definition 1.3,
respectively.
• For each of the BWAi bags-of-words, its measure of the cosine similarity distance
with all the BWA’ bag-of-words is calculated. This measurement allows us to
build a scoring matrix of the relationship identified at the activity level. If the
measure of similarity of a pair of the bags-of-words BWAi and BWA’i exceeds the
threshold U a , these are considered as message-type tasks. These tasks can be of
sub-type send or receive (BPMN 2.0 2011). Then, a vector that contains the pair
of message-type activities (PMA) that exceeded the threshold is generated.
1.4.3 Phase 3: Collaboration Discovery
1.4.3.1 Method 5: Business Process Discovery
The private business process model of each organization involved in inter-

organizational collaboration is discovered, from the event logs L and L’. In our
proposal, the split-miner algorithm presented in Augusto et al. (2019) is reused to
discover a business process model. This algorithm discovers the flow of the business
process, with its behavior through its different paths, decision points, bifurcations,
Another random document with
no related content on Scribd:
TO ROAST POTATO CARD
TIN-PAIL EAR
OVER OVEN GRATE SUPPORT
SCREEN-DOOR
EYE FOR FISH- SPRING REPAIR HOLDS BUGGY
HOOK SINKER CURTAIN OVER
SPRING HINGE FOR TORN BUTTON-
BOX HOLE
DRY-BATTERY CONNECTOR
These Eight Examples of How an Ordinary Safety Pin can be Used for
Practical Purposes Are Suggestive of Many Others
It is surprising to note how many uses can be made of an ordinary
safety pin, both as an emergency-repair device, and for other
purposes. I found 40 simple uses for safety pins, a few of which are
shown in the illustration. Some of the many other uses are as a
chain, a candle holder, a spring, a cover holder for a kettle, and as a
pencil clip. Most of these can be made by twisting the safety pin with
a small plier. By using a cutting plier, many more useful kinks can be
devised.—George G. McVicker, North Bend, Neb.
Moving Heavy Objects with a Broom
The Broom Is Slippery and Rides over the Floor or Even Carpets without
Trouble
After trying to move a heavy trunk alone, the attempt is usually

given up until some friend can be called in to lend a hand, because a
truck or other means of handling such heavy objects is not at hand.
An ordinary sweeping broom will serve as a sled to move a stove or
similar object in the home without heavy lifting. The straw in the
broom is slippery and can be drawn over carpets without injuring
them. In handling a stove, the legs must of course be removed, one
person drawing the load and the other steadying it.—Samuel H.
Avery, Chicago, Ill.
Rear Seat for Motorcycle or Bicycle
This Light-Weight Homemade Rear-Seat Fixture Is a Convenience for a

Strong Bicycle or Motorcycle
A rear seat mounted on a light support that can be quickly
attached to a strong bicycle or a motorcycle is handy, and one like
that shown in the sketch can be made in the home workshop. The
supporting frame, consisting of two main sections forming a fork over
the rear axle, and a brace extending to the bicycle frame, are made
of ¹⁄₄ by 1¹⁄₄-in. strap iron. The seat and the handlebars are
supported on posts of the usual type, flattened at their lower ends,
and riveted to the fixture. The lower ends of the fork are bent to form
foot rests. The detailed construction of the seat post and the method
of clamping the brace at A are shown in the smaller sketches. The
fastenings on the seat and handlebar posts are made with rivets or
bolts.—P. P. Avery, Garfield, N. J.
Changing Wheels Equalizes Wear on Baby-Cab
Tires
The front and rear wheels of a baby cab are usually of the same
size and the tires on the rear wheels are worn much more rapidly
than those on the front wheels. By changing the position of the
wheels, the wear on the tires is equalized, making it unnecessary to
renew them until all are worn out.—J. Cecil Alter, Cheyenne, Wyo.
A Craftsman Leather Billfold
By LIVINGSTON HAVILAND
T he making of a billfold is easily within the range of an amateur in

leather work, who will observe the instructions carefully. Seal,
morocco, pigskin, and ooze sheep are satisfactory. Skiver, or thin
leather, and silk moiré make suitable linings. When the processes
involved in the making of the billfold are mastered, numerous other
small articles in leather will suggest themselves, and can be made
similarly.
Fig. 1. Place the Pattern on the Leather and Trim It to Size Even with the
Edges of the Pattern
The finished billfold is shown with the cardcase side up; folded,
and with the bill compartment open, in Fig. 2. Begin the making of it
by cutting a pattern of stiff cardboard for the main portion, or body
part No. 1, as shown in Fig. 1.
Make a second pattern for body part No. 2, as shown in Fig. 1.
Trim the leather to fit the patterns, a straight edge of metal being
placed over the pattern as a guide for the knife. A close-grained
hardwood board, or a piece of sheet zinc, is suitable as a base for
cutting the leather.
Cut pieces of lining slightly larger than the patterns. If skiver is
used, it must be glued to the leather with a good quality of leather
glue, which should be pliable so as not to crack the leather. Do not
stretch the lining, but merely smooth it gently, as otherwise it may
cause the leather to buckle. Place the glued parts under a light
weight to dry. In fitting the silk lining into place, apply a thin stripe of
glue around the edges only, and permit them to extend beyond the
leather. Then trim them off neatly.
Part No. 1 is to be folded to form the back of the billfold and also
the two pockets for cards. The parts are glued as shown in Fig. 3.
Weight the billfold and permit the glue to dry. The sewing may then
be undertaken.
Fig. 2. The Photographs Show the Cardcase Side, the Folded Billfold, and
the Silk-Lined Money Compartment
Fig. 3. Glue the Parts Together Carefully along the Edges, and Fold the
Ends as Indicated
The neatness of the billfold will depend largely on the care with
which the sewing is done, and the finish on the edges. Draw a
straight line with a blunt-point tool along the edges to be sewed, ³⁄₁₆
in. from them. Mark the spacing for an inch of the stitches on a strip
of paper and transfer them to the stitching line. Punch the holes,
making certain that each is made with the awl held vertical, as
shown at D. Back the open ends with strips of cardboard when
punching holes in them. Heavy, waxed silk of a color to match the
leather is used for the sewing. The work may be done by hand
without a holder, clamped between two boards. Thread the two
needles and start them from the right side, as at E. Pass the needle
from the first hole through the second as at F. With the needles then
in position, as at G, continue this stitch. Tie the ends of the thread
neatly at the end of the stitching. Trim the edges and tool them to a
smooth finish with a suitable metal tool, or rod, heated moderately,
applying water with the finger tip.
¶The taper of a key is generally ¹⁄₈ in. per foot of length, and keys
having a head are preferable because of greater ease in removing
them. The width of a key is usually ¹⁄₄ the diameter of the shaft, plus
¹⁄₈ inch.
Lawn Mower Sharpened Efficiently with Simple
Rigging
The Lawn Mower is Sharpened Quickly by the Use of This Rigging
With this device, one can quickly sharpen the lawn mower, or
perhaps earn money sharpening the neighbors’ machines. Because
several knives in a lawn mower work against the cutting blade, it is
difficult to sharpen a mower satisfactorily with a file or stone. A usual
method is to reverse the cutting-wheel cogs, turn the mower upside
down, and run it backward over the ground. By the use of grinding
paste on the cutting blade the cutting wheel tends to sharpen itself.
This process can be improved by supporting the mower, as shown at
A, the blocks permitting the mower’s driving wheels to rest outside
them. A block, B, supports the wooden roller. Remove the driving
wheels, and the gears which turn the cutting wheel. Take the gear
from one end and put it in the other end, replacing the driving wheel.
By revolving the latter backward, the cutting wheel will also turn
backward. Apply a paste of emery powder and lubricating oil to the
cutting blade, and adjust the blade so that it comes fairly into contact
with the cutting wheel. Turn the driving wheel backward, and
gradually tighten the adjustment until a good edge on the cutting
blade and the knives is produced.
A handle fixed on the driving wheel, as shown at C, makes turning
it easier. After sharpening, replace the gears, clean the bearings
thoroughly, and replace the driving wheels.—Ed. M. Hawes, Everett,
Wash.
Rubber Pads for Opening Screw Watch Bezel
The modern screw case for watches is primarily for the purpose of
keeping dust from getting into the delicate works. Unless one is
possessed of a strong grip, it is hard not only to set the case tight
enough to keep out dust, but also to open the watch when
adjustment is necessary. To make this work easy, a jeweler has, as a
part of his tool kit, several rubber dies to fit the different sizes of
watch cases. He made them from old rubber heels, cut to fit the
watch cases closely. Danger of breaking the crystal is also
overcome.—C. E. Drayer, Cleveland, Ohio.
Spring Roost Releases Poultry-House Door Latch
The Door is Opened Automatically When the Fowl Leave Their Roost
Those in the habit of shutting their poultry houses at night to

prevent disturbance or loss by night-prowling animals, will be
interested in a contrivance that makes this unnecessary. The
principle of the mechanism, as illustrated is that the opening to the
poultry house is closed after the fowls have gone to roost, and in the
morning when they descend from the roost, the door automatically
opens, and they are free to go out. The spring door opens inside with
hinges at the side, and a suitable bumper. The spring has just
sufficient tension to cause the door to open when the catch is
released.
The roost is fitted up as shown in the detail. One end is held by a
bolt, while the other rests on a coil spring compressed by the weight
of the fowls on the roost. A cord is attached at the spring end,
passing through pulleys to the door, where it is fastened to the end of
the bolt catch. The door is closed after the fowls are on the roost.
The spring end is compressed to a level position, by their weight.
The instant the birds leave the roost, it rises and pulls the cord,
releasing the catch, and permitting the door to open.—George S.
Brown, Norwich, Conn.
A Child’s Bell-Ringing Hoop
In making this useful toy, first, a wheel from an old baby carriage
was stripped of the tire; then two laths are fastened together at one
end, and the wheel inserted between them, at the other end, and
held by a bolt. A strong cord was passed around the groove in the
wheel, and around a spool, nailed so as to turn on the side of the
lath when the wheel is rotated. Driven into the rim of the spool are
three nails, which strike against the old alarm-clock bell, fastened as
indicated.—W. F. MacGregor, Toronto, Can.

Download ebook New Perspectives On Enterprise Decision Making Applying Artificial Intelligence Techniques 1St Edition Julian Andres Zapata Cortes Editor Giner Alor Hernandez Editor Cuauhtemoc Sanchez Ramirez Editor online pdf all chapter docx epub

Uploaded by

Copyright:

Available Formats

You might also like

Download ebook New Perspectives On Enterprise Decision Making Applying Artificial Intelligence Techniques 1St Edition Julian Andres Zapata Cortes Editor Giner Alor Hernandez Editor Cuauhtemoc Sanchez Ramirez Editor online pdf all chapter docx epub

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Download ebook New Perspectives On Enterprise Decision Making Applying Artificial Intelligence Techniques 1St Edition Julian Andres Zapata Cortes Editor Giner Alor Hernandez Editor Cuauhtemoc Sanchez Ramirez Editor online pdf all chapter docx epub

Uploaded by

Copyright:

Available Formats

New Perspectives on Enterprise

Decision Making Applying Artificial

Handbook on Decision Making Volume 3 Trends and

Chess Explained The c3 Sicilian 1st Edition Sam Collins

Starting Out The c3 Sicilian 1st Edition John Emms

The Paper Issue 83 1st Edition Origamiusa

BNF 83 (British National Formulary) March 2022 Joint

English Grammar Exercises with answers Part 4 Your

Patriot vs Loyalist American Revolution 1775 83 Combat

British Light Infantryman vs Patriot Rifleman American

Julian Andres Zapata-Cortes

More information about this series at http://www.springer.com/series/7092

ISSN 1860-949X ISSN 1860-9503 (electronic)

Chapter 4, entitled “Map-Bot: Mapping Model of Indoor Work Environments in

Medellín, Colombia Julian Andres Zapata-Cortes

Part I Industrial Applications

7 ENERMONGRID: Intelligent Energy Monitoring,

Part II Decision-Making Systems for Industry

Part III Artificial Intelligence Techniques

Gustavo Alonso Acosta-Amaya Facultad de Ingeniería, Departamento de Instru-

Roberto Díaz-Reza Department of Electric Engineering and Computation, Univer-

Yara Anahí Jiménez-Nieto Faculty of Accounting and Administration, Univer-

Jezreel Mejía-Miranda Centro de Investigación en Matemáticas CIMAT, A.C,

Adolfo Rodríguez-Parada Faculty of Accounting and Administration, Univer-

Fig. 1.1 Model of the process choreography within a collaboration

Fig. 2.11 Comparison of Apriori, FPGrowth, PredictiveApriori

Fig. 4.9 Sketch of the Walling robot control structure. Source

Fig. 6.1 Proposed model and relationships among variables . . . . . . . . . . 126

Fig. 9.5 Layer of cognitive knowledge. Source The authors . . . . . . . . . . 209

Fig. 13.11 Questionnaire for knowledge validation in new income

Fig. 15.15 Results of decision trees algorithms for 50 queries and 4

Fig. 18.13 Representation of: A Kinetic energy; B Potential

Fig. 19.14 Panel D generational change . . . . . . . . . . . . . . . . . . . . . . . . . . . . 451

Table 1.1 Example of a case in the M-Parts participant’s event log . . . . 14

Table 3.6 Linguistic summaries - Truth value T vs. probability

Table 10.8 CFL rules generated using EUREKA UNIVERSE . . . . . . . . . 231

Table 15.6 Results of decision trees algorithms with 50 queries

Jaciel David Hernandez-Resendiz, Edgar Tello-Leal,

Abstract In an inter-organizational environment, the discovery of process choreog-

J. D. Hernandez-Resendiz · U. M. Ramirez-Alcocer · J. A. Mata-Torres

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 3

specification of message flows of the process choreography discovered, allowing to

Public and private organizations require proper knowledge asset management to

Fig. 1.1 Model of the

by sending and receiving messages. In a BPMN-based collaboration diagram, the

Therefore, inter-organizational process mining requires these historical data to be

1.2 Related Work

and unstructured attributes. Similarly, (Bala et al. 2018) propose a semi-automatic

1.3 Preliminaries and Definitions

Definition 1.3 (The bag-of-words of activities) Refers to a vector of words that

Definition 1.4 (Inter-organizational relationship) Indicates that two or more organi-

1.4.1 Phase 1: Processing of Event Logs

1.4.1.1 Method 1: Construction of Bags-of-Words

Fig. 1.2 Overview of the methodology approach

1.4.1.2 Method 2: Generation of the Scoring Matrix

1.4.2 Phase 2: Identifying the Correlation Between Events

1.4.2.1 Method 3: Selection at the Case Level

1.4.2.2 Method 4: Selection at the Activity Level

1.4.3 Phase 3: Collaboration Discovery