Artificial Intelligence and Industrial

Applications: Artificial Intelligence

Techniques for Cyber-Physical, Digital
Twin Systems and Engineering
Applications Tawfik Masrour
Visit to download the full and correct content document:
https://textbookfull.com/product/artificial-intelligence-and-industrial-applications-artifici
al-intelligence-techniques-for-cyber-physical-digital-twin-systems-and-engineering-ap
plications-tawfik-masrour/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Applications of Artificial Intelligence Techniques in

Engineering SIGMA 2018 Volume 1 Hasmat Malik

https://textbookfull.com/product/applications-of-artificial-
intelligence-techniques-in-engineering-
sigma-2018-volume-1-hasmat-malik/

Applications of Artificial Intelligence Techniques in

Engineering SIGMA 2018 Volume 2 Hasmat Malik

https://textbookfull.com/product/applications-of-artificial-
intelligence-techniques-in-engineering-
sigma-2018-volume-2-hasmat-malik/

Trends in Artificial Intelligence Theory and

Applications. Artificial Intelligence Practices: 33rd
International Conference on Industrial, Engineering and
Other Applications of Applied Intelligent Systems,
IEA/AIE 2020, Kitakyushu, Japan, September 22-25,
https://textbookfull.com/product/trends-in-artificial-
intelligence-theory-and-applications-artificial-intelligence-
Hamido Fujita
practices-33rd-international-conference-on-industrial-
engineering-and-other-applications-of-applied-intelligent-sys/

2020 International Conference on Applications and

Techniques in Cyber Intelligence: Applications and
Techniques in Cyber Intelligence (ATCI 2020) Jemal H.
Abawajy
https://textbookfull.com/product/2020-international-conference-
on-applications-and-techniques-in-cyber-intelligence-
applications-and-techniques-in-cyber-intelligence-
Artificial Intelligence Systems Based on Hybrid Neural
Networks: Theory and Applications Michael Zgurovsky

https://textbookfull.com/product/artificial-intelligence-systems-
based-on-hybrid-neural-networks-theory-and-applications-michael-
zgurovsky/

Artificial Intelligence for Marketing Practical

Applications 1st Edition Jim Sterne

https://textbookfull.com/product/artificial-intelligence-for-
marketing-practical-applications-1st-edition-jim-sterne/

Artificial Intelligence and Blockchain for Future

Cybersecurity Applications 1st Edition Yassine Maleh

https://textbookfull.com/product/artificial-intelligence-and-
blockchain-for-future-cybersecurity-applications-1st-edition-
yassine-maleh/

Artificial Intelligence for Big Data Complete guide to

automating Big Data solutions using Artificial
Intelligence techniques Anand Deshpande

https://textbookfull.com/product/artificial-intelligence-for-big-
data-complete-guide-to-automating-big-data-solutions-using-
artificial-intelligence-techniques-anand-deshpande/

Advances in Artificial Intelligence, Software and

Systems Engineering Tareq Z. Ahram

https://textbookfull.com/product/advances-in-artificial-
intelligence-software-and-systems-engineering-tareq-z-ahram/
Lecture Notes in Networks and Systems 144

Tawfik Masrour
Ibtissam El Hassani
Anass Cherrafi Editors

Artificial
Intelligence
and Industrial
Applications
Artificial Intelligence Techniques
for Cyber-Physical, Digital Twin
Systems and Engineering Applications
Lecture Notes in Networks and Systems

Volume 144

Series Editor
Janusz Kacprzyk, Systems Research Institute, Polish Academy of Sciences,
Warsaw, Poland

Advisory Editors
Fernando Gomide, Department of Computer Engineering and Automation—DCA,
School of Electrical and Computer Engineering—FEEC, University of Campinas—
UNICAMP, São Paulo, Brazil
Okyay Kaynak, Department of Electrical and Electronic Engineering
Bogazici University, Istanbul, Turkey
Derong Liu, Department of Electrical and Computer Engineering, University
of Illinois at Chicago, Chicago, USA, Institute of Automation, Chinese Academy
of Sciences, Beijing, China
Witold Pedrycz, Department of Electrical and Computer Engineering,
University of Alberta, Alberta, Canada, Systems Research Institute,
Polish Academy of Sciences, Warsaw, Poland
Marios M. Polycarpou, Department of Electrical and Computer Engineering,
KIOS Research Center for Intelligent Systems and Networks, University of Cyprus,
Nicosia, Cyprus
Imre J. Rudas, Óbuda University, Budapest, Hungary
Jun Wang, Department of Computer Science, City University of Hong Kong,
Kowloon, Hong Kong
The series “Lecture Notes in Networks and Systems” publishes the latest
developments in Networks and Systems—quickly, informally and with high quality.
Original research reported in proceedings and post-proceedings represents the core
of LNNS.
Volumes published in LNNS embrace all aspects and subfields of, as well as new
challenges in, Networks and Systems.
The series contains proceedings and edited volumes in systems and networks,
spanning the areas of Cyber-Physical Systems, Autonomous Systems, Sensor
Networks, Control Systems, Energy Systems, Automotive Systems, Biological
Systems, Vehicular Networking and Connected Vehicles, Aerospace Systems,
Automation, Manufacturing, Smart Grids, Nonlinear Systems, Power Systems,
Robotics, Social Systems, Economic Systems and other. Of particular value to both
the contributors and the readership are the short publication timeframe and the
world-wide distribution and exposure which enable both a wide and rapid
dissemination of research output.
The series covers the theory, applications, and perspectives on the state of the art
and future developments relevant to systems and networks, decision making, control,
complex processes and related areas, as embedded in the fields of interdisciplinary
and applied sciences, engineering, computer science, physics, economics, social, and
life sciences, as well as the paradigms and methodologies behind them.
** Indexing: The books of this series are submitted to ISI Proceedings,
SCOPUS, Google Scholar and Springerlink **

More information about this series at http://www.springer.com/series/15179

Tawfik Masrour Ibtissam El Hassani
• •

Anass Cherrafi
Editors

Artificial Intelligence
and Industrial Applications
Artificial Intelligence Techniques
for Cyber-Physical, Digital Twin Systems
and Engineering Applications

123
Editors
Tawfik Masrour Ibtissam El Hassani
Department of Mathematics Department of Industrial
and Computer Science and Manufacturing Engineering
National Graduate School National Graduate School
for Arts and Crafts for Arts and Crafts
Meknes, Morocco Meknes, Morocco

Anass Cherrafi
Department of Industrial
and Manufacturing Engineering
National Graduate School
for Arts and Crafts
Meknes, Morocco

ISSN 2367-3370 ISSN 2367-3389 (electronic)

Lecture Notes in Networks and Systems
ISBN 978-3-030-53969-6 ISBN 978-3-030-53970-2 (eBook)
https://doi.org/10.1007/978-3-030-53970-2
© Springer Nature Switzerland AG 2021
This work is subject to copyright. All rights are reserved 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

It is fairly obvious that our world is uncertain, complex, and ambiguous. It is true
today more than ever, and tomorrow will be more challenging. Artificial intelli-
gence (AI) can improve responses to major challenges to our communities from
economic and industrial development to health care and disease containment.
Nevertheless, artificial intelligence is still growing and improving technology, and
there is a necessity for more creative studies and researches led by both academics
and practitioners.
Artificial Intelligence and Industrial Applications—A2IA’2020, which is the first
edition of an annual international conference organized by the ENSAM—Meknes
at Moulay Ismail University, intends to contribute to this common great goal. It
aims to offer a platform for experts, researchers, academics, and industrial practi-
tioners working in artificial intelligence and its different applications to discuss
problems and solutions, concepts, theories and map out the directions for future
research. The connections between institutions and individuals working in this field
have to keep growing on and on, and this must have a positive impact on pro-
ductivity and effectiveness of researches.
The main topics of the conference were:
– Smart Operation Management
– Artificial Intelligence: Algorithms and Techniques
– Artificial Intelligence for Information and System Security in Industry
– Artificial Intelligence for Energy
– Artificial Intelligence for Agriculture
– Artificial Intelligence for Health care
– Other Applications of Artificial Intelligence
In A2IA’2020 conference proceedings, about 141 papers were received from
around the world. A total of 58 papers are selected for presentation and publication.
In order to maintain a high level of quality, a blind peer review process was
performed by a large international panel of qualified experts in the conference topic
areas. Each submission received at least two reviews, and several received up to

v
vi Preface

five. The papers are evaluated on their relevance to A2IA’2020 tracks and topics,
scientific correctness, and clarity of presentation.
The papers are organized in two parts:
– Artificial Intelligence and Industrial Applications:
Smart Operation Management (Volume 1)
In: Advances in Intelligent Systems and Computing
– Artificial Intelligence and Industrial Applications:
Artificial Intelligence Techniques for Cyber-Physical, Digital Twin Systems
and Engineering Applications (Volume 2)
In: Lecture Notes in Networks and Systems
We hope that our readers will discover valuable new ideas and insights.
Lastly, we would like to express our thanks to all contributors in this book
including those whose papers were not included. We would also like to extend our
thanks to all members of the Program Committee and reviewers, who helped us
with their expertise and valuable time. We are tremendously grateful for the pro-
fessional and organizational support from Moulay Ismail University. Finally, our
heartfelt thanks go especially to Springer Nature.

Tawfik Masrour
Ibtissam El Hassani
Anass Cherrafi
Organization

General Chair

Tawfik Masrour Department of Mathematics and Computer

Science, Artificial Intelligence for Engineering
Sciences Team (IASI), Laboratory of
Mathematical Modeling, Simulation and
Smart Systems (L2M3S), ENSAM, My Ismail
University, 50500 Meknes, Morocco
t.masrour@ensam.umi.ac.ma
t.masrour@umi.ac.ma

Co-chair
Vincenzo Piuri Department of Computer Science University
of Milan via Celoria 18, 20133 Milano (MI),
Italy

Keynotes Speakers
Amal El Fallah Seghrouchni Sorbonne University, Paris, France
Mawa Chafii ENSEA, CY Paris University, Paris, France
Abdellatif Benabdellah University of Le Havre, France
Ali Siadat Arts et Métiers Paris Tech Metz, France
Jiju Antony Heriot-Watt University, Edinburgh, Scotland
Andrew Kusiak University of Iowa, USA

vii
viii
TPC Chairs
Alexandre Dolgui, France
Jose Arturo Garza-Reyes, UK
Abby Ghobadian, UK
Kannan Govindan, Denmark
Janusz Kacprzyk, Poland
Vikas Kumar, UK
Ali Siadat, France
International Scientific Committee
Aalaoui Zinab, Morocco
Abawajy Jemal H., Australia
Aboulaich Rajae, Morocco
Aghezzaf El-Houssaine, Belgium
Ahmadi Abdeslam, Morocco
Ait Moussa Abdelaziz, Morocco
Akhrif Iatimad, Morocco
Aksasse Brahim, Morocco
Al-Mubaid Hisham, USA
Alali Abdelhakim, Morocco
Alami Hassani Aicha, Morocco
Ali Ahad, USA
Ali Siadat, France
Allahverdi Ali, Kuwait
Aly Ayman A., Saudi Arabia
Arbaoui Abdelaziz, Morocco
Azizi Abdelmalek, Morocco
Azzeddine Mazroui, Morocco
Babai Mohamed Zied, France
Badie Kambiz, Iran
Balas Valentina Emilia, Romania
Bansal Jagdish Chand, India
Batouche Mohamed Chawki,
Saudi Arabia
Behja Hicham, Morocco
Belhadi Amine, Morocco
Ben Abdllah Mohammed, Morocco
Benabbou Rajaa, Morocco
Benaissa Mounir, Tunisia
Benghabrit Asmaa, Morocco
Benoussa Rachid, Morocco
Berrada Mohamed, Morocco
Bouhaddou Imane, Morocco
Organization
Brouri Adil, Morocco
Buccafurri Francesco, Italy
Carrabs Francesco, Italy
Castillo Oscar, Mexico
Cerulli Raffaele, Italy
Chaouni Benabdellah Abla, Morocco
Chaouni Benabdellah Naoual, Morocco
Charkaoui Abdelkabir, Morocco
Chbihi Louhdi Mohammed Reda,
Morocco
Cherrafi Anass, Morocco
Ciaramella Angelo, Italy
Ciasullo Maria Vincenza, Italy
D’Ambrosio Ciriaco, Italy
Daoudi El-Mostafa, Morocco
De Mello Rodrigo Fernandes, Brazil
Deep Kusum, India
Dolgui Alexandre, France
Ducange Pietro, Italy
El Akili Charaf, Morocco
El Haddadi Anass, Morocco
El Hammoumi Mohammed, Morocco
El Hassani Ibtissam, Morocco
El Jai Mustapha, Morocco
El Jasouli Sidi Yasser, Belgium
El Mazzroui Abb Elaziz, Morocco
El Mghouchi Youness, Morocco
El Ossmani Mustapha, Morocco
Elbaz Jamal, Morocco
Elfezazi Said, Morocco
Elmazroui Azz Addin, Morocco
Es-Sbai Najia, Morocco
Ettifouri El Hassane, Morocco
Organization
Ezziyyani Mostafa, Morocco
Faquir Sanaa, Morocco
Fassi Fihri Abdelkader
Fiore Ugo, Italy
Fouad Mohammed Amine, Morocco
Gabli Mohamed, Morocco
Gaga Ahmed, Morocco
Gao Xiao-Zhi, Finland
Garza-Reyes Jose Arturo, UK
Ghobadian Abby, UK
Giuseppe Stecca, Italy
Govindan Kannan, Denmark
Grabot Bernard, France
Hajji Tarik, Morocco
Hamzane Ibrahim, Morocco
Harchli Fidaa, Morocco
Hasnaoui Moulay Lahcen, Morocco
Herrera-Viedma Enrique, Spain
Itahriouan Zakaria, Morocco
Jaara El Miloud, Morocco
Jaouad Kharbach, Morocco
Jawab Fouad, Morocco
Kacprzyk Janusz, Poland
Kaya Sid Ali Kamel, Morocco
Khadija Bouzaachane, Morocco
Khireddine Mohamed Salah, Morocco
Khrbach Jawad, Morocco
Kodad Mohssin, Morocco
Krause Paul, UK
Kumar Vikas, UK
Laaroussi Ahmed, Morocco
Lagrioui Ahmed, Morocco
Lasri Larbi, Morocco
Lazraq Aziz, Morocco
Lebbar Maria, Morocco
Leung Henry, Canada
Manssouri Imad, Morocco
Marcelloni Francesco, Italy
Massoud Hassania, Morocco
Medarhri Ibtissam, Morocco
ix
Mkhida Abdelhak, Morocco
Mohiuddin Muhammad, Canada
Moumen Aniss, Morocco
Moussi Mohamed, Morocco
Najib Khalid, Morocco
Nee Andrew Y. C., Morocco
Nfaoui Elhabib, Morocco
Nguyen Ngoc Thanh, Poland
Nouari Mohammed, France
Noureddine Boutammachte, Morocco
Novák Vilém, Czech
Ouazzani Jamil, Morocco
Ouerdi Noura, Morocco
Oztemel Ercan, Turkey
Palmieri Francesco, Italy
Pesch Erwin, Germany
Pincemin Sandrine, France
Rachidi Youssef, Morocco
Rahmani Amir Masoud, Iran
Raiconi Andrea, Italy
Rocha-Lona Luis, Mexico
Saadi Adil, Morocco
Sabor Jalal, Morocco
Sachenko Anatoliy, Ukraine
Sael Nawal, Morocco
Saidou Noureddine, Morocco
Sekkat Souhail, Morocco
Senhaji Salwa, Morocco
Serrhini Simohammed, Morocco
Sheta Alaa, USA
Siarry Patrick, France
Soulhi Aziz, Morocco
Staiano Antonino, Italy
Tahiri Ahmed, Morocco
Tarnowska Katarzyna, USA
Tyshchenko Oleksii K., Czech
Tzung-Pei Hong, Taiwan
Zemmouri Elmoukhtar, Morocco
Zéraï Mourad, Tunisia
x Organization

Local Organizing Committee

Abou El Majd Badr, Med V University
Ahmadi Abdessalam, ENSAM
Benabbou Rajaa, ENSEM
Benghabrit Asmaa, ENSMR
Benghabrit Youssef, ENSAM
Bouayad Aboubakr, ENSAM
Bouhaddou Imane, ENSAM
Chaira Abdellatif, Fac Sci, Meknes
Cherrafi Anass, ENSAM
El Hassani Ibtissam, ENSAM
Hajji Tarij, ENSAM
Masrour Tawfik, ENSAM
Najib Khalid, ENSMR
Saadi Adil, ENSAM
Sekkat Souhail, ENSAM
Zemmouri El Moukhtar, ENSAM
Publication Chairs
Badr Abou El Majd, Morocco
Ibtissam El Hassani, Morocco
Ercan Oztemel, Turkey
Cherrafi Anass, ENSAM
El Hassani Ibtissam, ENSAM
Hajji Tarij, ENSAM
Masrour Tawfik, ENSAM
Najib Khalid, ENSMR
Saadi Adil, ENSAM
Sekkat Souhail, ENSAM
Zemmouri El Moukhtar, ENSAM
Poster Chairs
Benabbou Rajaa, Morocco
Souhail Sekkat, Morocco
Registration Chairs
Abdessalam Ahmadi, Morocco
Adil Saadi, Morocco
Jalal Sabor, Morocco
Web Chairs
Ibtissam El Hassani, Morocco
Zemmouri El Moukhtar, Morocco
Organization xi

Public Relations Chairs

Abla Benabdellah Chaouni, Morocco
Asmaa Benghabrit, Morocco
Youssef Benghabrit, Morocco
Imane Bouhaddou, Morocco
Said Ettaqi, Morocco
Industrial Session Chairs
Anass Cherrafi, Morocco
Souhail Sekkat, Morocco
Ph.D. Organizing Committee
Amhraoui ElMehdi, Morocco
Benabdellah Chaouni Abla, Morocco
Eddamiri Siham, Morocco
El Mazgualdi Choumicha, Morocco
El Mekaoui Fatima Zahra, Morocco
Fattah Zakaria, Morocco
Hadi Hajar, Morocco
Jounaidi Ilyass, Morocco
Khdoudi Abdelmoula, Morocco
Moufaddal Meryam, Morocco
Raaidi Safaa, Morocco
Rhazzaf Mohamed, Morocco
Zekhnini Kamar, Morocco
Contents

Efficient Pore Network Extraction Method Based

on the Distance Transform . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Adam Hammoumi, Maxime Moreaud, Elsa Jolimaitre, Thibaud Chevalier,
Alexey Novikov, and Michaela Klotz
A Study of Energy Reduction Strategies in Renewable Hybrid Grid . . . 14
Chahinaze Ameur, Sanaa Faquir, and Ali Yahyaouy
DFS Response Time Prediction Using the Techniques
of “Deep Learning” . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
Akram Elomari, Larbi Hassouni, and Abderrahim Maizate
Spam Filtering System Based on Nearest Neighbor Algorithms . . . . . . . 36
Ghizlane Hnini, Jamal Riffi, Mohamed Adnane Mahraz, Ali Yahyaouy,
and Hamid Tairi
A Robust Control of Permanent Magnet Synchronous Generator
Based Wind Energy Conversion System via Online-Tuned Artificial
Neural Network Compensators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
Mohsin Beniysa, Aziz El Janati El Idrissi, Adel Bouajaj,
and Mohammed Réda Britel
Classification and Watermarking of Brain Tumor Using Artificial
and Convolutional Neural Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Youssef Douzi, Tarik Hajji, Mohammed Benabdellah, Abdelmalek Azizi,
and Tawfik Masrour
Machine Learning Application for Blood Pressure Telemonitoring
over Wireless Sensors Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
A. El Attaoui, S. Largo, A. Jilbab, and A. Bourouhou

xiii
xiv Contents

TRNSYS Simulation of a Solar Cooling System Under

Oujda Climate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
Sara El Hassani, Hanane Ait Lahoussine, Hamid Amaoui,
Mohammed Amine Moussaoui, and Ahmed Mezrhab
Fuzzy Logic Based Intelligent Vertical Handover Decision
in Heterogeneous Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
Saida Driouache, Najib Naja, and Abdellah Jamali
Toward Intelligent Solution to Identify Learner Attitude
from Source Code . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
Zakaria Itahriouan, Nisserine El Bahri, Samir Brahim Belhaouari,
Hajji Tarik, and Mohamed Ouazzani Jamil
A New Method to Analysis of Internet of Things Malware Using
Image Texture Component and Machine Learning Techniques . . . . . . . 119
Saloua Senhaji, Sanaa Faquir, Fidae Harchli, Hajji Tarik,
and Mohammed Ouazzani Jamil
Optimization of the SLM Process by Printing a Test Tray
in AlSi7Mg06 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
Faraj Zainab, Aboussaleh Mohamed, and Zaki Smail
Mining Online Opinions and Reviews Using Bi-LSTM
for Reputation Generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
Achraf Boumhidi, Abdessamad Benlahbib, and El Habib Nfaoui
Theme Identification for Linked Medical Data . . . . . . . . . . . . . . . . . . . . 145
Siham Eddamiri, Elmoukhtar Zemmouri, and Asmaa Benghabrit
A Proposal for a Deep Learning Model to Enhance Student Guidance
and Reduce Dropout . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
Mouhcine Sabri, Jaber El Bouhdidi, and Mohamed Yassin Chkouri
EduBot: An Unsupervised Domain-Specific Chatbot
for Educational Institutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166
Md. Abdullah Al Muid, Md. Masum Reza, Ruhullah Bin Kalim,
Nayeem Ahmed, Md. Tarek Habib, and Md. Sadekur Rahman
SQL Generation from Natural Language Using Supervised Learning
and Recurrent Neural Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175
Youssef Mellah, El Hassane Ettifouri, Abdelkader Rhouati,
Walid Dahhane, Toumi Bouchentouf, and Mohammed Ghaouth Belkasmi
An Intelligent Power Distribution Management with Dynamic
Selection in Smart Building Based on Prosumers Classification
and an Intelligent Controller . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184
Yassin Bendriss, Youssef Hamdaoui, and Fatima Guerouate
Contents xv

Conceptual Architecture of AI-Enabled IoT System for Knee

Rehabilitation Exercises Telemonitoring . . . . . . . . . . . . . . . . . . . . . . . . . 200
Mohamed El Fezazi, Atman Jbari, and Abdelilah Jilbab
A WBAN Platform for Healthcare: Real-Time Remote Monitoring
of Human Vital Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
Elhoussaine Baba, Abdelilah Jilbab, and Ahmed Hammouch
Energy Management Strategy of a Fuel-Cell Electric Vehicle
Based on Wavelet Transform . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220
Nada Rifai, Jalal Sabor, and Chakib Alaoui
Natural Language Processing: Challenges and Future Directions . . . . . 236
Zakaria Kaddari, Youssef Mellah, Jamal Berrich,
Mohammed G. Belkasmi, and Toumi Bouchentouf
A Genetic Algorithm Based Optimal Sizing Strategy
for PV/Battery/Hydrogen Hybrid System . . . . . . . . . . . . . . . . . . . . . . . . 247
Nouhaila Lazaar, Eyman Fakhri, Mahmoud Barakat, Jalal Sabor,
and Hamid Gualous
Data Mining Model for Student Internship Placement Using Modified
Case Based Reasoning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 260
Ferddie Quiroz Canlas
PIL Implementation of Adaptive Gain Sliding Mode Observer
and ANN for SOC Estimation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 270
Yahia Mazzi, Hicham Ben Sassi, Fatima Errahimi, and Najia Es-Sbai
Enhancing the Energy Performance of Passive Building Through
the Internet of Things . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279
El Mehdi El Khattabi, Omar Diouri, Mohamed Mharzi,
and Mohammed Ouazzani Jamil
Modelling and Optimization of Stirling Engine for Waste Heat
Recovery from Cement Plant Based on Adiabatic Model
and Genetics Algorithms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287
Kaoutar Laazaar and Noureddine Boutammachte
Embedded Skin Cancer Detection and Classification
on Raspberry Pi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297
Khihel Ibrahim, Youssef Filali, My Abdelouahed Sabri,
and Abdellah Aarab
Design Approach for Assembly Lines Reconfiguration Process . . . . . . . 307
Zakaria Moukkar, Aline Cauvin, and Alain Ferrarini

Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 323

About the Editors

Dr. Tawfik Masrour is a professor of Applied

Mathematics and Artificial Intelligence in National
High School for the Arts and Crafts
(ENSAM-Meknes), My Ismaïl University UMI, and
member of Research Team Artificial Intelligence for
Engineering Sciences (AIES) and Laboratory of
Mathematical Modeling, Simulation and Smart
Systems. (L2M3S). He graduated from
Mohammed V University – Rabat with MSc degree
in Applied Mathematics and, Numerical Analysis, and
from Jacques-Louis Lions Laboratory, Pierre and
Marie Curie University, Paris with M.A.S (DEA) in
Applied Mathematics, Numerical Analysis and
Computer Science. He obtained his PhD, in
Mathematics and informatics, from École des Ponts
ParisTech (ENPC), Paris, France. His research inter-
ests include Control Theory and Artificial Intelligence.
Email : t.masrour@umi.ac.ma

xvii
xviii About the Editors

Dr. Ibtissam El Hassani is a professor of Industrial

and Logistic Engineering in National High School for
the Arts and Crafts (ENSAM-Meknès), My Ismaïl
University UMI, and head of Research Team Artificial
Intelligence for Engineering Sciences (AIES). She
graduated as an industrial engineer and then got her
Ph.D. in Computer Engineering & Systems
Supervision at the Center for Doctoral Studies at My
Ismail University. Her fields of interest are applications
of artificial intelligence in industrial engineering spe-
cially in Lean Manufacturing, Six Sigma &
Continuous Improvement, Production systems,
Quality, Management System, Health and security at
work, Supply Chain Management, Industrial Risk
Management, and Work Measure.
Email: i.elhassani@umi.ac.ma

Dr. Anass Cherrafi Is an Assistant Professor at the

Department of Industrial Engineering, ENSAM-
Meknes, Moulay Ismail University, Morocco. He
holds a PhD in Industrial Engineering. He has about
7 years of industry and teaching experience. He has
published a number of articles in leading international
journals and conferences. He has participated as Guest
Editor for special issues in various international
journals. His research interests include industry 4.0,
green manufacturing, Lean Six Sigma, integrated
management systems and supply chain management.
 Efficient Pore Network Extraction
Method Based on the Distance Transform

Adam Hammoumi1(B) , Maxime Moreaud1 , Elsa Jolimaitre1 ,

Thibaud Chevalier2 , Alexey Novikov3 , and Michaela Klotz3
1
IFP Energies nouvelles, Rond-point de l’échangeur de Solaize,
BP 3, 69360 Solaize, France
adam.hammoumi@ifpen.fr
2
IFP Energies nouvelles, 1 et 4 avenue de Bois-Préau,
92852 Rueil-Malmaison, France
3
LSFC Laboratoire de synthèse et fonctionnalisation des céramiques UMR 3080
CNRS/SAINT-GOBAIN CREE, Saint-Gobain Research Provence, 550 avenue
Alphonse Jauffret, Cavaillon, France

Abstract. Digital twins of materials allow to achieve accurate predic-

tions that help creating novel and tailor-made materials with higher stan-
dards. In this paper, we are interested in the characterization of porous
media. Our attention is drawn to develop a method to describe accu-
rately the pore network microstructure of porous materials as presented
in [7]. This work proposes an efficient algorithm based on the distance
transform method [12] which is a widely used method in image pro-
cessing. The followed approach suggests that a distance transform map,
obtained from a microstructure image, passes through different steps.
Starting from local maxima extraction and filtering operation, to end up
with another distance transform with source propagation. We illustrate
our algorithm with the well-known Pore Network Model of the litera-
ture [13], which supposes that the pore structure is either a network
of connected cylinders or cylinders and spheres. Our approach is also
applied on multi-scale Boolean random models modelling complex porous
media microstructures [11]. The porous media morphological character-
istics extracted could be used to simulate complex phenomena as the
physisorption isotherms or other experimental techniques.

Keywords: Distance transform · Local maxima · Porous material ·

Multi-scale Boolean model · Pore network model · Physisorption ·
Experimental techniques

1 Introduction
Modeling physico-chemical phenomena in porous materials is a topic of great
interest. It allows the study of real materials but also the design of new ones.
Methods for flow simulation inside a porous media have been widely investigated
in the last decades. The accuracy of these methods depends on a wide vari-
ety of factors that can be categorized into two families: physical and chemical
c Springer Nature Switzerland AG 2021
T. Masrour et al. (Eds.): A2IA 2020, LNNS 144, pp. 1–13, 2021.
https://doi.org/10.1007/978-3-030-53970-2_1
2 A. Hammoumi et al.

properties, such as permeability, capillary pressure; and geometrical-topological

properties, in particular surface area and interconnectivity of pores. Regardless
any physical consideration, the extraction of internal geometry and topology of
considered microstructures is a fundamental step before any numerical simula-
tion. However, this process is still ambiguous. We cite from [1] obstacles to this
approach, such as the lack of concise definitions of what constitutes pore and
pore throats in real material with complex geometries. That may have dramatic
effect on the network topology. According to the same reference, the uniqueness
of numerical simulation results is not guaranteed for different network models of
the same material. In fact, in [4], it was demonstrated that different lattice-based
topologies of the same data sample produced different water retention results.
A rough validation of phenomena as the mechanisms of adsorption/desorption
can be obtained from an over-simplified geometries, such as the ink bottle pore
illustrated in [8]. Another widely known method is the pore network model [9]
which is used to simulate physical and chemical processes, in particular phase
exchange and non-Darcy flow. The pore network in this type of representations is
a discrete grid where the nodes represent the pores and the connections between
them constitute the pore necks. These elements are usually represented by sim-
ple geometrical elements like spheres or cylinders. The challenge for this type of
approach is not the representation of the porous network but rather its adap-
tation to fit realistic microstructures in terms of simulation results. Another
alternative is multi-scale stochastic morphological models detailed in [10]. These
models can be used to predict effective material properties of complex geometries
[22]. The modelization of the microstructure relies on Boolean random models
based on stochastic point processes [2,3,11]. A typical microstructure obtained
by this method contains elements like spheres or platelets [22]. Increasingly com-
plex microstructures can be produced by combining different Boolean models;
a multi-scale microstructure is generated by performing basic set operations -
union and intersection - to several one scale microstructures. In this configura-
tion, the solid space is represented by the precedent elements and the porosity
space is deduced from the remaining space.
A fundamental advantage of digital pore network generation methods is com-
putation efficiency. A valid formulation of a typical pore network extraction prob-
lem is the one that asks: how to discretize the void space into a series of pores and
pore necks? In this context, we should differentiate between computer generated
microstructures and real microstructures obtained by acquisition, namely X-ray
tomography. We also keep in mind that the nature of the materials under study
is closely linked to the modeling method. For example, the network extraction
algorithm developed in [5] is unsuitable for highly porous materials. The aim
is not to cite all the techniques of pore extraction. Instead we mention some
of the closest procedures to our proposal. The first example is skeletonization-
based methods, in which the estimation of the medial axe leads to the creation
of a graph of pore space. In this approach, pores are placed at the nodes of
the skeleton and pore–necks are defined by the curved elements of the skeleton
that connect these nodes [6]. Another example is the marker-based watershed
Efficient Pore Network Extraction Method Based on the Distance Transform 3

segmentation that proceeds, after applying a distance transform to the void

space, to remove all local maxima that lead to an oversegmented watershed [7].
In this work, a distance transform based algorithm is proposed. We extract from
the distance map local maxima. A filtering operation is then applied to avoid
overlapping issues. Then, we cover the void space by the geodesic distance trans-
form that yields then again a discretized space of pores. Our approach takes as
an input a binary image that represents a porous material microstructure. The
binary aspect of the image refers to the co-existence of both solid and porosity
space. Our algorithm is called the Pore Network Partitioning method (PNP) and
it aims at treating a wide spectrum of microstructures.

2 Method
In this part, for the sake of simplicity, equations and illustrations are mainly
provided for the 2D case. Associated results can be found in Sect. 3.

2.1 Distance Transform

Consider a 2D Image I represented in the discrete plane E = Z × Z. Each point
P (pixel) is identified by the pair (xp , yp ). I is divided into two subsets. The
first one is the reference set of points which contains foreground elements. The
second one is its complementary. A distance map is an image that evaluates each
element of I by its distance from the closest point belonging to the reference set.
The distance transform is the operation that computes the distance map. It is
the basis of many image processing methods [19]. This transformation relies on a
specifc metric d. We distinguish between exact methods based on the euclidean
distance [14,16] and approximate methods based on the chamfer distances [15].
The euclidean distance for two points P and Q of the discrete space Z2 is given
by: 
de (P, Q) = (xP − xQ )2 + (yP − yQ )2 (1)
In the following, a point will be denoted only by one letter in 2D or 3D. The
euclidean distance is the exact distance. In Z3 , the equation is given similarly
by: 
de (P, Q) = (xP − xQ )2 + (yP − yQ )2 + (zP − zQ )2 (2)
However, it can be costly in terms of time complexity. The chamfer distances
were introduced to accelerate the computation [15]. They evaluate the distance
between two points as the shortest path according to a pixel-connectivity rela-
tion. The chamfer distance between two points x and y can be expressed as:

r
dc (x, y) = min (Wi dti (xi , xi−1 )) (3)
i

where dti is the distance between two neighboring points xi and xi−1 . r is the
index of the target point y. The two approaches differ in terms of the propagation
4 A. Hammoumi et al.

technique considering the weights Wi attributed to local displacements for the

chamfer distances. That is, the distance transform is a local transformation. The
operator of the distance transform reads:

DT d (x) = min d(x, y) x, y ∈ E (4)

y=P

where the related n-dimensional metric space is denoted (E, d), X is the solid
space and P = E \ X is the porosity space. Given a square orthogonal grid in
E, the used algorithm throughout this paper is the raster scanning algorithm
which is well established in the literature [16]. It consists of a forward and a
backward pass. The two passes algorithm ensures obtaining the correct distance
to the nearest foreground elements. This technique is widely used for the problem
of connected component labeling [17]. Algorithms based on distance transform
can produce Voronoi Diagrams and Delaunay triangulation [18]. In Fig. (1-a), we
show side by side the original image and its corresponding distance transform. We
obtain a digital representation of the porosity space, where the intensity of the
white color of background elements is inversely proportional to their proximity
to foreground elements. The distance transform can also yield a skeletonization
transform – an operation in image processing that simplifies an object while
retaining its topology – [20]. The accuracy of the distance transform depends on
whether we use exact or approximate transformations.

2.2 Local Maxima Extraction

To reconstruct the porosity network, we are interested in the extraction of the

farthest points from the foreground elements. These points are extracted and
stored. We call this operation Local Maxima Extraction. Each element (pixel or
voxel) value will be compared with other elements of its vicinity. The elements
of maximum value are the ones of interest. The output is therefore sensitive to
the number of the neighbors. The corresponding operator writes as follows:

θ(DT d (x)) = DT d (xk ) (5)
xk

for point xk that verify:

DT d (xk ) ≥ DT d (xk ) (6)
Where point xk belongs to the neighborhood V(xk ).
In Fig. (1-b), the dots correspond to the local extracted maxima.

2.3 Maxima Filtering

For each maximum point, we create a disk (or a sphere in 3D case) parame-
terized by Sxrkk ⊂ P , where the radius is the corresponding distance transform
rk = DT d (xk ). This step will cover the areas around the extracted maxima.
Included elements will have all the same distance transform value. Following the
Efficient Pore Network Extraction Method Based on the Distance Transform 5

(a) (b)

(c)

(d)

Fig. 1. Illustrations of intermediate results of the PNP Method applied to a 2D image

of a Boolean model of spheres [22]. (a) Distance transform. (b) Maxima extraction
from the obtained distance map. (c) Maxima expansion before filtering. (d) Maxima
expansion after filtering. Each figure contains the initial image on the left and its
transform on the right. Solid space is identified by white color in the original image
and by black color in the transforms. The pore space is initially black and is identified
by gray-scale colors when it’s discretized. In the (d) transform, a part of pore network
and solid space are both black.
6 A. Hammoumi et al.

creation of disks around maxima points, intermediate disks can be trapped in-
between other disks. In this case, we check if there’s an overlapping between the
intermediate disk center and its vicinity. Then, at the same position, we compare
the distance transform value and the value of the other disks. If the center value
is lower, then the disk is useless and the center corresponding to the maximum
is removed. We write:
∀xk Sxrkk ≥ DT d (xk ) =⇒ xk = 0 (7)
A subsequent operation of creating disks around the remaining points is then
applied. In Fig. (2), the center of the black disk is included into a bigger disk.
Applying the filter will remove the associated maximum point and its disk
accordingly. Considering intersection of disks, two filtering tecnhiques are intro-
duced: standard filtering, shown in Fig. (2-a) and filtering with intersections
removal, shown in Fig. (2-b). We define the partition function M and F for each
filtering technique:
⎧
⎪
⎨rk x ∈ (∃ !Sxk )
rk

M (x) = rk x ∈ A| A = i Sxrii , rk ≥ ri (8)

⎪
⎩
0 otherwise

rk x ∈ (∃ !Sxrkk )
F (x) = (9)
0 otherwise
M and F refer to standard filtering and filtering with intersections removal
respectively. For M , we keep the intersections between pores. Whereas, for the
other technique, we create intermediate pores by removing the intersections.
The interest of these filtering methods will be explored further in what follows.
Figures (1-c) and (d) illustrate the obtained results before and after the standard
filtering operation. For the 3D case, all the operations described above remain
the same, replacing disks with spheres.

Standard filtering

Before After

(a)
Filtering with intersections removal

Before After

(b)

Fig. 2. Illustrations of the local maxima filtering techniques for the 2D case. (a) Stan-
dard filtering. (b) Filtering with intersections removal.
Efficient Pore Network Extraction Method Based on the Distance Transform 7

2.4 Distance Transform with Source Propagation

Previous steps do not yet allow to obtain a partition, some points of the porosity
being unlabeled. A geodesic distance, as illustrated in Fig. (3), measures the
length of the path between two points, this path being constrained to be included
in a given set [21]. The geodesic distance between a point x of P and a subset
X of P is denoted:
DG (x, X) = inf

DG (x, x ) (10)
x ∈X

Fig. 3. The geodesic distance between P and Q is given by the red path, whereas the
Euclidean distance is the length of the blue path.

Creation of disks aims not to cover all the porosity space. Filtering operation
with intersections removal creates void as well in-between disks. Therefore, we
use the geodesic distance transform to fill empty spaces. In parallel with the
distance calculation which is carried out step by step, it is possible to propagate
the initial source point. We write:

Sp (x, X) = x | inf

DG (x, x ) (11)
x ∈X

The actual propagation of the geodesic distance transforms starts from the
previously created disks and does not end until all the space is completely filled
in. In that manner, we insure covering all the void space. Figures (4-c) and (d)
illustrate the pore network partition on a 2D Boolean model of spheres obtained
by the two former filtering techniques. The shapes of pores can be controlled by
altering the used technique. Standard filtering (keeping the intersections) aims
to obtain rounded pore shapes shown in Figs. (4-a) and (c). And filtering with
intersections removal produces voronoi-like pore partitions shown in Figs. (4-b)
and (d).
8 A. Hammoumi et al.

Fig. 4. Illustrations of the two types of local maxima filtering applied to a 2D image of
a Boolean model spheres. (a) standard filtering and (c) its corresponding pore network.
(b) filtering with intersections removal and (d) its corresponding pore network.

3 Discussion

The method is evaluated in terms of: accessibility of pores, realistic modeling of

the porous network and time complexity. Our procedure allows the extraction of
the porosity network in few simple steps. The algorithm has no preliminary infor-
mation about the spatial organization of the pores inside the porous space, but
will gather information through a series of image transformations. This method
yield a labeled image, where each pore is identified by its size. The accessibil-
ity of pores –which is mandatory for accurate simulations– is guaranteed. The
propagation of the geodesic distance transform permits probing pores with arbi-
trary shapes. Realistic pore networks are then modeled. Since we use only two
distance transforms throughout the procedure, the time complexity of both used
algorithms is linear time O(N ) [12]. Therefore, the method generates good inputs
for numerical simulations using simple image transformations at the least com-
putation cost. The following pseudo code summarizes the main steps explained
before (cf. Algorithm 1).
Our algorithm allows pore network extraction on images of 2003 voxels in
1.6 s and bigger images of 5003 in 29 s. The computations were performed using
a personal computer (CPU: intel core i7 2.6 GHz, RAM: 16 GB). For the sake of
Efficient Pore Network Extraction Method Based on the Distance Transform 9

Algorithm 1. PNP method applied to a 2D microstructure image

1: procedure Pore Network Partitioning
2: I ← Binary Image
3: Definition of W, H as the width and the height of I
4: T ←W ×H
5: for n ∈ [0, T ] do
6: Computation of :
7: Distance map DT d (n);  Distance Transform
8: Maxima map M d (n);  Maxima Extraction
9: end for
10: for n ∈ [0, T ] do  Preliminary Maxima Expansion
11: if M d (n) = 0 then
12: r ← DT d (n)
13: Create Disk S(n, r);
14: end if
15: end for
16: for n ∈ [0, T ] do  Maxima Standard Filtering
17: if M d (n) = 0 then
18: r ← DT d (n)
19: if S(n, r) ≥ DT d (n) then
20: M d (j, i) = 0
21: end if
22: end if
23: end for
24: Repeat Steps 10 → 15  Final Maxima Expansion
25: for n ∈ [0, T ] do
26: Computation of :
27: Geodesic Distance map DG (n);  Pore Network reconstruction
28: end for
29: end procedure

comparison, the marker-based watershed segmentation algorithm [7] took 200 s

for similar sized images of 5003 voxels with the same kind of personal computer.
We provide our results for two types of micro-structures. The first one is
a 2D pore network model taken from [13]. It is made of interconnected cylin-
ders assembling pores and pores necks. The parameterization of this network
is relatively simple. Each cylinder is identified by its length and diameter. The
intersections of cylinder’s endpoints constitute the pore network. In Fig. (5-b),
we show an example of 8002 pixels image.
A multi-scale microstructure is also being considered. It is based on the multi-
scale cox Boolean models [11]. To generate this type of microstructure, we fix the
simulation parameters for the aggregates and for the grains; size of aggregates,
size of grains and grain volume fraction Vv . The multi-scale aspect is being
defined by: the aggregates, the grains inside and the outer space. We study the
case of a multi-scale Boolean model of platelets represented in a volume of 2003
voxels. A platelet is defined by its length, height and thickness (L = 25, H =
15, T = 5), the rest of the parameters are: Vvin = 0.2, Vvout = 0.4, Rinc = 100
10 A. Hammoumi et al.

(a) (b)

Fig. 5. Illustration of the PNP method applied to a 2D image of pore network model.
(a) The original microstructure (white: solid space, black: porosity space). (b) Corre-
sponding partition pore network (grayscale elements: pores, black elements: solid).

(radius of the sphere modeling the aggregates), Vvinc = 0.4. The PNP relative
result is shown in Fig. (6). Partition labeling (originated from a mapping from
the distance transform to a greyscale image) is provided.

Partition label
39
36
35
12
10

5
3
0

(a) (b)

Fig. 6. Illustrations of the extracted pore networks by the PNP method. (a) 3D Pore
network of a multi-scale Boolean model of platelets. (b) Section of the pore network
(a) with partition labeling.

A porous material is characterized by descriptors as the volume fraction of

pores, surface area, pore size distribution, etc. These parameters are obtained
experimentally but could also be simulated numerically. This method should be
Efficient Pore Network Extraction Method Based on the Distance Transform 11

·106 PSD
6
Gr

5
Cumulative sum [Number of voxels]

0 5 10 15 20 25 30 35 40
Pore size [voxels]

Fig. 7. Pore Size Distribution of the microstructure shown in Fig. (6-a) (complementary
of multi-scale Boolean model of platelets).

regarded as a first brick for numerical simulations of porous material properties.

For instance, it can be used to compute the pore size distribution.
Consider the family of sets (each set is made of pores), and every set has a
unique label (pore diameter  DT d ). We consider the family of sets {Lr } such
that r = 1, 2, ....n and n = max(DT d ). A Granulometry function is obtained
with:
Gr = Cardinal(Lr ) (12)
The pore size distribution ‘PSD’ is the cumulative sum of the granulometry
function values and is defined by:
H0 = 0
where r = 0, 1, ...n − 1 (13)
H r+1 = Gr+1 + H r
An example of the PSD computation is shown in Fig. (7).
The interested reader can find the PNP algorithm on the open access software
“plug im!” website, https://www.plugim.fr
The Figs. (1), (4), (5), (6) were generated using “plug im!” (2020).

4 Conclusion
A new algorithm for pore network extraction has been described. The algorithm
relies on simple and well-known methods in image processing. The method is
12 A. Hammoumi et al.

a mapping from the continuum void space contained initially in a computer-

generated microstructure to a discretized space made of distinct pores in terms
of size and shape. In particular, the method is suitable for complex microstruc-
tures as shown above, but also for discrete grids made of interconnected spheres
and/or cylinders. At this stage, the method is potentially interesting due to its
simplicity and computational efficiency. The labeled extracted pore network fits
numerical simulations requirements. It makes it possible to manage pores and
their connectivities. Future work will focus on enhancing the accuracy of the
proposed algorithm. Namely, by the use of efficient neural network algorithms.
The simulation of physico-chemical processes based on the data produced by the
PNP algorithm is the next step to validate the method.

References
1. Bhattad, P., Willson, C.S., Thompson, K.E.: Effect of network structure on char-
acterization and flow modeling using x-ray micro-tomography images of granular
and fibrous porous media. Transp. Porous Med. 90, 363 (2011)
2. Serra, J.: Image Analysis and Mathematical Morphology. Academic Press, London
(1982)
3. Matheron, G.: Random Sets and Integral Geometry. Wiley, New York (1975)
4. Vogel, H.J.: A numerical experiment on pore size, pore connectivity, water reten-
tion, permeability, and solute transport using network models. Eur. J. Soil Sci.
51(1), 99–105 (2000)
5. Yang, F., Hingerl, F., Xiao, X., et al.: Extraction of pore-morphology and capillary
pressure curves of porous media from synchrotron-based tomography data. Sci.
Rep. 5, 10635 (2015)
6. Pudney, C.: Distance-based skeletonization of 3D images. In: Proceedings of the
1996 IEEE TENCON Digital Signal Processing Applications, TENCON 1996, vol.
201, pp. 209–214 (1996)
7. Gostick, J.T.: Versatile and efficient pore network extraction method using marker-
based watershed segmentation. Phys. Rev. E 96, 023307 (2017)
8. Zeng, Y., Fan, C., Do, D.D., Nicholson, D.: Evaporation from an ink-bottle pore:
mechanisms of adsorption and desorption. Ind. Eng. Chem. Res. 53(40), 15467–
15474 (2014)
9. Xiong, Q., Baychev, T.G., Jivkov, A.P.: Review of pore network modelling of
porous media: experimental characterisations, network constructions and appli-
cations to reactive transport. J. Contam. Hydrol. 192, 101–117 (2016)
10. Jeulin, D.: Morphology and effective properties of multi-scale random sets: a
review. Comptes Rendus Mécanique 340(4–5), 219–229 (2012)
11. Moreaud, M., Chaniot, J., Fournel, T., Becker, J.M., Sorbier, L.: Multi-scale
stochastic morphological models for 3D complex microstructures. In: IEEE Con-
ference on 17th Workshop on Information Optics (WIO), Quebec city, Canada
(2018)
12. Maurer Jr., C.R., Qi, R., Raghavan, V.: A linear time algorithm for computing
exact Euclidean distance transforms of binary images in arbitrary dimensions.
IEEE Trans. Pattern Anal. Mach. Intell. 25(2), 265–270 (2003)
13. Ledezma, G., Verstraete, J., Sorbier, L., Leinekugel-Le Cocq, D., Jolimatre, E.,
Jallut, C.: Computational characterization techniques applied to a pore network
model by using a fast percolation algorithm (article under preparation, 2020)
Another random document with
no related content on Scribd:
 The importance of this invention can hardly be overestimated. It
ranks with Maudslay’s slide-rest and the turret tool-holder, as it is an
essential feature in all modern automatic lathes, both for bar-stock
and chucking work.
Assured of the success of the machine, Spencer withdrew from
active connection with the Billings & Spencer Company in 1874, and
in 1876, with George A. Fairfield, then superintendent of the Weed
Sewing Machine Company, and others, formed the Hartford Machine
Screw Company, one of the most successful enterprises in the city.
Unfortunately, Mr. Spencer withdrew in 1882 to manufacture a new
repeating shotgun and rifle which he had invented. The gun was a
success mechanically, but the Spencer Arms Company, which had
been formed in 1883 at Windsor, Conn., was a failure, and Mr.
Spencer lost heavily. In his later years Mr. Spencer has returned to
the field where he did his most brilliant work, automatic lathes. He
represents the New England mechanic at his best, and his tireless
and productive ingenuity has made a permanent impress on modern
manufacturing methods.
Francis A. Pratt was born at Woodstock, Vt. When he was eight
years old his family moved to Lowell. He was a mechanic from
boyhood but he had the good fortune to be apprenticed as a
machinist with Warren Aldrich, a good mechanic and a wise teacher.
At twenty, Mr. Pratt went to Gloucester, N. J., where he was
employed first as a journeyman, later as a contractor. In 1852 he
came to the Colt shop, where he worked for two years. He then
accepted the foremanship of the Phœnix Iron Works, which was run
by Levi Lincoln and his two sons.
Amos Whitney was born in Maine and moved to Lawrence, Mass.,
where he served his apprenticeship with the Essex Machine
Company which built cotton machinery, locomotives and machine
tools. He came from a family of mechanics. His father was a
locksmith and machinist, his grandfather was an expert blacksmith,
his great-grandfather was a small manufacturer of agricultural tools,
and he is of the same family as Eli Whitney of New Haven, and
Baxter D. Whitney, the veteran tool builder of Winchendon. In 1850
both he and his father were working at Colt’s factory at Hartford. In
1854 Amos Whitney joined Pratt in the Phœnix Iron Works, where
they worked together for ten years, the former as a contractor, the
latter as superintendent. Whitney was earning over eight dollars a
day when he left Colt’s and took up the new contract work which
offered at the beginning only two dollars a day.
Many of the shops of that generation were “contract shops.” The
Colt Armory was run on that basis, at least in its manufacturing
departments. Under this system the firm or company furnished all
the materials, machinery, tools, shop room and supplies, while the
workmen were employed by the contractor, their wages being paid
by the firm but charged against the contractor’s account. A better
training for future manufacturers could hardly be devised, and a
surprising number of these old-time contractors have succeeded
later in business for themselves.
In the summer of 1860 Pratt and Whitney rented a small room
and, in addition to their regular employment, began doing work on
their own account, i.e., manufacturing the small winder for the
Willimantic Linen Company. Mr. Whitney’s father-in-law acted as
pattern maker, millwright, bookkeeper and general utility man. The
following February they were burned out, but were running again a
month later in other quarters. Here they continued to spread from
room to room until all available space was outgrown. They
succeeded from the very start, and at once became leaders and
teachers of other mechanics, suggesters of new methods of work
and of new means for its accomplishment. Both Pratt and Whitney
were thoroughly familiar with gun manufacture, and the business
was hardly started when the outbreak of the Civil War gave them
more than they could do. In 1862 they took into partnership Monroe
Stannard of New Britain, each of the three contributing $1200. Mr.
Stannard took charge of the shop, as Pratt and Whitney were still
with the Phœnix Iron Works. Within two years the business had
increased to such an extent that they gave up their positions at the
Phœnix works and in 1865 erected the first building on their present
site. From $3600 in 1862 their net assets grew in four years to
$75,000, and during the three years following that they earned and
put back into the business more than $100,000. In 1869 the Pratt &
Whitney Company was formed with a capital of $350,000, later
increased to $500,000. In 1893 it was reorganized with a
capitalization of $3,000,000. Since that time it has become a part of
the Niles-Bement-Pond Company.
Figure 35. Francis A. Pratt
 Figure 36. Amos Whitney

Beginning with the manufacture of machine tools and tools for

making guns and sewing machines, they have extended their lines
until their catalog fills hundreds of pages. From their wide experience
in interchangeable manufacture, it was natural that they should take
a prominent part in developing the machinery for the manufacture of
bicycles and typewriters, when, later, these were introduced.
Soon after the Franco-Prussian War, an agent of the company
visited Prussia and found the royal and private gun factories
equipped with old and inferior machinery and the armories bare. Mr.
Pratt was sent for, and returned to Hartford with orders from the
German Government for $350,000 worth of gun machinery. During
the next few years Mr. Pratt made no less than ten trips to Europe,
taking orders aggregating over $2,000,000 worth of machinery.
When the panic of 1873 prostrated the industries of the United
States, Pratt & Whitney had orders, mostly foreign, which kept them
busy until 1875. Their equipment of the three royal armories of
Spandau, Erfurt and Danzig resulted in an improvement in quality of
output and a saving of 50 per cent in wages. Pratt & Whitney’s
production of gun-making machinery alone has run into many
millions of dollars, and there are few governments which have not at
one time or another purchased from them.
Pratt & Whitney from the start were leaders in establishing
standards, particularly in screw threads. Their gauges for pipe
threads have for years been the standard for the country. The
troubles which arose from the lack of agreement of standard gauges
and the growing demand for interchangeable bolts and nuts led to a
demand on the company for a set of gauges upon which all could
agree.
In undertaking this work Pratt & Whitney examined their own
standards of length with reference to government and other
standards in this country and abroad. The results were conflicting
and very unsatisfactory. By different measurements the same bar
would be reported as above and as below the standard length, and
the investigation produced no results which could be used for a
working basis. At length Prof. William A. Rogers of Harvard
University, and George M. Bond, backed by the Pratt & Whitney
Company, developed the Rogers-Bond comparator with which they
determined the length of the standard foot. When they began, the
length of the yard and its subdivisions varied with the number of
yardsticks. Professor Rogers’ work was based on line measurement
rather than the end measurement which had held sway from the time
of Whitworth and which is now generally recognized to be inferior for
final reference work. Professor Rogers went back of all the
secondary standards to the Imperial Yard in London and the
standard meter in the Archives at Paris. He obtained reliable
transfers of these, and with the coöperation of the United States
Coast Survey, the most delicate and exhaustive comparisons were
made of the standard bars prepared by him for the use of the
company with the government standard yard designated “Bronze No.
11.” Many thousands of dollars and three years of time went into this
work.
The methods used and the results obtained were examined and
reported upon by a committee of the American Society of
Mechanical Engineers, and the conclusion given in their report is as
follows:
The completion of the Rogers-Bond comparator marks a long stride in advance
over any method hitherto in use for comparison and subdivision of line-measure
standards, combining, as it does, all the approved methods of former observers
with others original with the designers. Comparisons can thus be checked
thoroughly by different systems, so that the final result of the series may be relied
on as being much nearer absolute accuracy than any hitherto produced.
The calipering attachment to the comparator deserves special commendation,
being simple in the extreme, and solving completely the problem of end
measurements within the limit of accuracy attainable in line reading, by means of
the microscope with the micrometer eye-piece. The standard to which the end
measurements are referred is not touched, and each measurement is referred
back to the same zero, so that error from end wear does not enter into the
problem. This attachment is in advance of all hitherto known methods of
comparing end measures, either with other end measures or with line standards,
both as to rapidity of manipulation and accuracy of its readings, the strong point in
its construction being that it refers all end measures to a carefully divided and
investigated standard bar, which is not touched during its use, and cannot be in the
slightest degree injured by this service, thus giving convincing assurance that the
measures and gauges produced by its use will be accurate and interchangeable.
 In the opinion of this committee, the degree of accuracy already attained is such
that no future improvements can occasion changes sufficiently great to affect the
practical usefulness of the magnitudes here determined, or the interchangeability
of structures based upon them with those involving further refinement.
Prof. W. A. Rogers and Mr. George M. Bond are unquestionably entitled to great
credit for the admirable manner in which they have solved the problem of exact
and uniform measurement, while the enterprise of the Pratt & Whitney Company in
bringing the whole matter into practical shape, is deserving of the thanks of the
engineering community.[183]
[183] Those interested may find detailed descriptions of the methods used
and of the Rogers-Bond comparator in the following references: George M.
Bond: Paper on “Standard Measurements,” Trans. A. S. M. E., Vol. II, p. 80.
George M. Bond: Paper on “A Standard Gauge System,” Trans. A. S. M. E.,
Vol. III, p. 122. Report of Committee on Standards and Gauges, Trans. A. S.
M. E., Vol. IV, p. 21 (quoted above). W. A. Rogers: Paper, “On a Practical
Solution of the Perfect Screw Problem,” Trans. A. S. M. E., Vol. V, p. 216.
Two lectures delivered by George M. Bond before the Franklin Institute,
Philadelphia, in 1884, on: 1. “Standards of Length and their Subdivision.” 2.
“Standards of Length as Applied to Gauge Dimensions.”

The standards so obtained became the basis of the gauges which

Pratt & Whitney have produced.
In 1888 the company received its first order for Hotchkiss revolving
cannon, and for three- and six-pounders rapid-fire guns. They have
made hundreds of these guns for the secondary batteries of war
vessels. In 1895 they brought out a one-pounder invented by E. G.
Parkhurst, an expert mechanic, who had entered their employment
as assistant superintendent in 1869 and later took charge of their
gun department.
For many years the Pratt & Whitney tool-room lathes were the
standard for the country. Later their leadership was materially
affected by the Hendey Machine Company of Torrington, Conn., who
built a high grade tool-room lathe having the change-gear box which
has since been applied to nearly all types of machine tools. The
change-gear box is one of the important contributions to tool building
made in recent years. Among the later developments introduced by
Pratt & Whitney is the process of thread milling, and they have
designed a full line of machines for this work.
 The Pratt & Whitney works, like the Colt Armory, has been a
training school for successful tool builders. Worcester R. Warner and
Ambrose Swasey were both foremen at Pratt & Whitney’s and left
there to go west, first to Chicago and then to Cleveland. Some
account of these two men will be given in a later chapter. Pratt &
Whitney have had a marked influence on tool building in Cleveland,
for, in addition to Warner and Swasey, E. C. Henn and Hakewessel
of the National Acme Manufacturing Company who developed the
multi-spindle automatic lathe, A. F. Foote of Foote, Burt & Company,
and George C. Bardons of Bardons & Oliver, come from their shop.
Johnston of Potter & Johnston, Pawtucket, was chief draftsman at
Pratt & Whitney’s; and J. N. Lapointe who later developed the
broaching machine, Dudley Seymour of Chicago, Gleason of the
Gleason Works in Rochester, E. P. Bullard of Bridgeport, and F. N.
Gardner of Beloit, Wis., inventor of the Gardner grinder, were all
workmen there.
Mr. Gleason was also a workman in the Colt Armory. He went to
Rochester in 1865 and the works which he developed form the most
important tool building interest in western New York. There have
been “a good many starts there in tool building and almost as many
finishes.” Mr. Gleason always said that but for the training and
methods he had gained at Hartford he would have shared their fate.
Like many others, his company began with a general line of machine
tools but has come to specialize on one type of machine, bevel-gear
cutters, of which they build a most refined type.
E. P. Bullard, like Gleason, worked at both Colt’s and Pratt &
Whitney’s. Later he formed a partnership with J. H. Prest and William
Parsons, manufacturing millwork and “all kinds of tools” in Hartford.
In 1866 he organized the Norwalk Iron Works Company of Norwalk,
but afterwards withdrew and continued the business in Hartford. For
a number of years Mr. Bullard was in the South and Middle West, at
Athens, Ga., at Cincinnati, where he organized the machine tool
department of Post & Company, and at Columbus, where he was
superintendent of the Gill Car Works. In 1875 he established a
machinery business in Beekman Street, New York, under the firm
name of Allis, Bullard & Company. Mr. Allis withdrew in 1877 and the
Bullard Machine Company was organized. Recognizing a demand
for a high grade lathe he went to Bridgeport, Conn., and engaged A.
D. Laws to manufacture lathes of his design, agreeing to take his
entire output. In the latter part of the same year Mr. Bullard took over
the business and it became the Bridgeport Machine Tool Works. In
1883 he designed his first vertical boring and turning mill, a single
head, belt feed machine of 37 inches capacity. This is believed to be
the first small boring machine designed to do the accurate work
previously performed on the face plate of a lathe. Up to that time
boring machines were relied on only for large and rough work. In
1889 he transferred his New York connections to J. J. McCabe and
gave his entire attention to manufacturing, the business being
incorporated as the Bullard Machine Tool Company in 1894.
The building of boring mills gradually crowded out the lathes, and
for twenty years the company has concentrated on the boring
machine as a specialty. In their hands it has received a remarkable
development. They introduced a range of small-sized mills capable
of much more accurate work than had been done on this type of
machine. They applied the turret principle to the head carried on the
cross rail and a few years later introduced a mill having a head
carried on the side of the frame which permitted of the close working
of the tools. These improvements transformed the boring mill into a
manufacturing machine, and it became practically a vertical turret
lathe with the advantages inherent in that type of machine. This
trend toward the lathe type has finally resulted in a multiple station-
type of machine which is in effect a vertical multi-spindle automatic
chucking lathe with five independent tool heads, as shown in Fig. 56.
Comparison of this with Fig. 15, shows how the lathe has developed
in the 115 years since Maudslay introduced the slide-rest principle
and the lead screw.
 CHAPTER XV
ROBBINS & LAWRENCE
A glance at the genealogical chart, Fig. 37, will show why the old
Robbins & Lawrence shop, at Windsor, Vt., in the backwoods of
northern New England, deserves a special chapter. When built, it
was miles away from a railroad. It was never large, and the wheels
of the original shop have long since ceased to turn, but few plants
have had so great an influence on American manufacturing. Three
brilliant mechanics, Lawrence, Howe and Stone, were working there
in the early fifties, and from them and their successors came wholly,
or in part, the vertical lathe turret, the miller, the profiler and a large
number of the modern machines used in interchangeable
manufacture. Of these three, Lawrence went to Hartford, Howe to
Providence, while Stone remained at Windsor. In each case an
important line of influence may be traced.
In the region about Windsor, sixty or seventy years ago, there
were a number of small custom gun shops, and one firm, N. Kendall
& Company, was regularly making guns at the Windsor prison, using
prison labor in addition to that of a number of free mechanics, who
did the finer work. The history of the Robbins & Lawrence Company
begins about 1838, when Lawrence came to Windsor from the
neighborhood of Watertown, N. Y. Fortunately he wrote out an
account of his life shortly before his death, at the request of his son,
giving a very interesting record of his early work and his connections
with his various manufacturing enterprises. This account shows
clearly the integrity, modesty and worth of the man.[184]
[184] By the courtesy of Mr. Ned Lawrence this account is given in
Appendix A. It has never been published before.
 N. KENDALL & CO. R. S LAWRENCE
Kendall & Lawrence
Custom Gun Shop, Windsor, Vt.
ROBBINS & LAWRENCE
Guns and Gun Machinery, Turret Lathes, Millers, etc.
R S. Lawrence, H. D. Stone, F. W. Howe
ENFIELD CHAS. E.
GUN BILLINGS
MACHRY., LAMSON, GOODNOW & Billings &
1855 YALE, 1859 Spencer,
Enfield, later Hartford
England E. G. LAMSON & CO.
SHARPS Guns, Sewing Machines, PROVIDENCE
J. R. BROWN
RIFLE Machine Tools, TOOL
S. B.
WORKS Windsor WORKS
DARLING,
Hartford, F. W. Howe,
ETC.
Conn. Supt., 1853-68
J. D. ALVORD
a contractor in R.
& L. Shop,
Hartford & Sharpe
Wks. Built the
Wheeler & Wilson
 Shop,
Bridgeport
BROWN &
SHARPE
WEED SEWING F. W. Howe,
MACH. CO. Supt., 1868-73
Hartford, Conn. Plain and Univ.
Sewing Machine business sold Millers, Turret
about 1861 to Mr. White Lathes
POPE MFG. RHODE ISLAND TOOL CO.
CO. Successors of Providence Tool
Hartford, Works
Conn.
WHEELER & PUTNAM
WILSON MACH. CO.
Bridgeport, Conn. Fitchburg,
Sewing Machines Mass.
S. C. Wright
WINDSOR MANUFACTURING SULLIVAN MACHRY.
WHITE
CO. CO.
SEWING
1865 Claremont, N. H.
MACH. CO.
Mine and Quarrying
Cleveland, O.
Machinery
CLEVELAND AUTO JONES, LAMSON & CO., JAMES
MACH. CO., ETC. 1869 HARTNESS
Cleveland, O. JONES & LAMSON 1889
WINDSOR MACHINE CO., 1879 FITCHBURG
MACHINE CO. Moved from Windsor to MACH. WORKS
Windsor, Vt., Springfield, Vt., in 1889 Fitchburg, Mass.
1889, Hartness Flat Turret, Fay Lo-Swing Lathe
Gridley Automatic Lathes
Automation
JAMES HARTNESS
BRYANT
E. R. FELLOWS FELLOWS GEAR SHAPER
CHUCKING
G. O. GRIDLEY CO.
GRINDER CO.
Wm. BRYANT Springfield, Vt.
Springfield, Vt.

Figure 37. Genealogy of the Robbins & Lawrence Shop

Richard S. Lawrence, whose portrait appears in Fig. 40, was born

in Chester, Vt., in 1817. When two years old, his father moved to
Jefferson County, N. Y., and his boyhood was spent in the
neighborhood of Watertown. He was only nine years old when his
father died, and consequently he had a hard boyhood, with very little
schooling, and was early at work in the support of the family. He
worked on a farm and later in a woodworking shop, making
carpenter’s and joiner’s tools. In the basement of this place was a
custom gun shop, where he spent much of his spare time and
became an expert gun maker. He worked with indifferent success at
various jobs until the winter of 1837-1838, when he served in the
United States army for three months, guarding the frontier during the
Canadian Rebellion. At his discharge he determined to start in
elsewhere for himself and thought of his relatives in Vermont. After a
long journey by the Erie Canal, the newly built Albany &
Schenectady Railroad, and by stage, he reached Windsor in 1838.
A week or two after his arrival, while visiting a Doctor Story, he
undertook to repair an old rifle, a “Turkey rifle,” made by the doctor’s
brother in a gun shop in the neighborhood, and put on a peep-sight,
a thing never heard of before in that neighborhood. He took the gun
apart, leaded out the barrel, forged and finished the sight and put it
on the gun. His skill in handling tools astonished those who watched
him. Two days later, when the work was done, the doctor and
Lawrence went out to try the gun. They paced off twelve rods from a
maple tree which had a three-quarter-inch auger hole in it that had
been used for a sap spout. Lawrence did the shooting. His own
account of it is as follows: “The doctor tended target. Could find no
ball hole. Said I missed the tree. I fired again, no ball hole to be
found. Doctor came up to me and said I had spoiled his rifle. Before
my repairs he could kill a chicken every time at twelve rods. I said,
‘Uncle, I am very sorry, but I will make the gun all right before I leave
it.’ He said he could not consent to my doing anything more to
improve the shooting qualities—the sight he liked very much. I said
that as the gun was loaded I would take one more shot and see if I
could not hit the tree. After the third shot I went up to the tree to
investigate, and all of the three balls which I had fired were found in
the auger hole.”[185] The doctor was astonished, for he had never
heard of such shooting.
[185] Quoted from the full account given in Appendix A.
 The next day he took Lawrence down to see N. Kendall &
Company, who were making guns at the Windsor prison. They hired
him at once for two years at $100 a year. His first work was stocking
rifles by hand and the first day he put on five stocks. The next day
the superintendent looked over the work and said it was well done,
but it would never do to rush the work as he had, for he would “soon
gun-stock them out of town,” and he “must take it more easy.” In the
course of the next six months, he had so far mastered every process
in the factory, even that of engraving in which he could soon
compete with the oldest hands, that he was put in charge of the
shop. Four years later the company gave up the gun business, and
for a time Lawrence remained as foreman of the carriage department
in the prison shop.
In 1843 Kendall and Lawrence hired a small shop in Windsor
village and started a custom gun shop. In the winter of 1844 S. E.
Robbins, a business man, came to them and said that the
Government was in the market for 10,000 rifles. The matter was
talked over, a partnership formed, and a bid sent to Washington. In
spite of the opposition of nearly all the other Government
contractors, who said they could never do the work, it resulted in the
award of a contract for 10,000 to Robbins, Kendall & Lawrence, at
$10.90 each, attachments extra, to be furnished within three years.
They bought land, built a shop, and bought or made the necessary
machinery. It was in the performance of this and the subsequent
contract that many of the early machine tools were developed. The
contract was finished eighteen months ahead of time, at a good
profit, and they obtained a second contract for 15,000 at the same
price. Soon after finishing the first contract, Robbins and Lawrence
bought out Kendall’s interest in the firm, which became Robbins &
Lawrence. The business proved very profitable. About 38 per cent of
their work for the Government had to be rejected on account of poor
material and workmanship, but the California gold excitement was
then at its height and guns were in great demand. They were
therefore able to sell their second-quality work for the full
government price. About 1850 they contracted with Courtland C.
Palmer for 5000 Jennings rifles, a gun which later developed through
the Henry rifle into the present well-known Winchester rifle.
 About 1850 Robbins & Lawrence took the first of the steps which
led to their undoing. The railroad had just been completed through
Windsor, and S. F. Belknap, a large railroad contractor, induced them
to start in the car business, which, of course, had no rational relation
with their main activity of building guns. Mr. Belknap assured them
that he could control all the car work in that section, and put in
$20,000 as a silent partner. The firm went to a large outlay, but just
as they were finishing the first cars, Belknap quarreled with the
president of the railroad and the firm could not sell a single car when
they had expected to. After a considerable delay they were sold to
other roads, and stock which proved valueless was taken in
payment. The operation involved an actual loss of $134,000, which
was later increased to nearly $240,000.

Figure 38. Robbins & Lawrence Armory, Windsor, Vt.

From an Old Lithograph

In all of their gun work, Robbins & Lawrence used the

interchangeable system, and they contributed very largely to its
development. Lawrence, Howe, and later Stone, were constantly
improving the methods of manufacture. Fitch’s article on
Interchangeable Manufacture in the U. S. Census Report of 1880,
describes and illustrates a profiling machine built by Howe as early
as 1848. The design shown there was used for many years
throughout all the gun shops in the country. He also designed a
barrel drilling and rifling machine, and he and Lawrence designed
and built a plain miller, which was the forerunner of the well-known
Lincoln miller. One of these millers, built in 1853, is still running in the
shop of the North Brothers Manufacturing Company in Philadelphia.
This machine had a rack-and-pinion feed for the table, which
chattered badly when starting a heavy cut. The principal
improvement which F. A. Pratt introduced in the Lincoln miller was
the substitution for this of a screw and nut. The original drawing of
this Robbins & Lawrence machine is still on file in the office of the
Jones & Lamson Machine Company and shows clearly that it
furnished the basis of the design of the Lincoln miller.
In 1851 Robbins & Lawrence sent to the Exposition in London a
set of rifles built on the interchangeable system, which excited great
interest and for which they received a medal. This led to the visit of
an English commission which resulted in a large contract to Robbins
& Lawrence for Enfield rifles, and for gun machinery which was
installed in the Armory at Enfield, near London. It has been said that
this contract caused the failure of Robbins & Lawrence. This is not
true.
In 1852 the company contracted to make 5000 Sharps carbines at
Windsor, and 15,000 rifles and carbines at a plant which they were to
erect in Hartford. The Sharps Company advanced $40,000 to enable
them to build a new factory and Mr. Lawrence moved to Hartford in
1853 to superintend the building and equipment of the plant. Shortly
after it was completed, Robbins & Lawrence, already strained by
their losses in the car-building venture and with the erecting of the
new plant, undertook a contract with Fox, Henderson & Company for
25,000 Minié rifles. They were assured by the agent that he had in
his pockets contracts for 300,000 more, which he promised them on
the completion of the 25,000. Lawrence objected strenuously to
signing the contract for the 25,000 without more assurance as to the
300,000 to follow, as the outlay for the work would greatly exceed
the profits on the first contract. It was signed, however, and it later
developed that the agent had no authorization for the 300,000. It was
this which caused the failure of Robbins & Lawrence.
Mr. Lawrence left the firm and took charge of the new Hartford
plant which had been bought by the Sharps Rifle Company. J. D.
Alvord, one of the contractors at Hartford under Lawrence, later built
the Wheeler & Wilson plant at Bridgeport. Robbins and others leased
the Windsor shops and began the manufacture of sewing machines.
In 1859 the plant and business were purchased by Lamson,
Goodnow & Yale, who retained Henry D. Stone as their mechanical
expert. During the Civil War the plant was given over entirely to the
manufacture of army rifles, and the sewing-machine business was
sold to Mr. White of the White Sewing Machine Company of
Cleveland, Ohio.
In the early thirties Silas Lamson had begun manufacturing scythe
snaths in one of the hill towns of western Massachusetts. Up to that
time the farmers had either used straight poles or those which
happened naturally to have a convenient twist. Lamson conceived
the idea of steaming the poles and bending them to a predetermined
curve. About 1840 his sons, Nathan and E. G. Lamson, moved to
Shelburne Falls and after some years began the manufacture of
cutlery, founding the factory which has been in successful operation
ever since. After the completion of the railroad through Windsor, they
moved their snath factory to that place. They and their successors,
the Lamson & Goodnow Manufacturing Company, continued this
work there for many years. When the Robbins & Lawrence property
was put on the market it was purchased by E. G. Lamson, A. F.
Goodnow and B. B. Yale, under the name of Lamson, Goodnow &
Yale. E. G. Lamson & Company and the Windsor Manufacturing
Company succeeded this firm and continued the manufacture of
machine tools and Ball and Palmer carbines, and completed a
number of government rifle contracts. In 1869 R. L. Jones, a
business man, of the Ascutney Mill at Windsor, joined the firm, which
became Jones, Lamson & Company, and a small cotton mill was
added to their other activities. Ten years later the Jones & Lamson
Company was organized to take over the machine business. During
all these changes Henry D. Stone continued as the designer. A large
poster of the Windsor Manufacturing Company, printed some time
about 1865, shows that they had plenty of irons in the fire, for they
were prepared to furnish guns and machinery for manufacturing
guns, sewing machines and needles, a standard line of hand-
operated turret lathes, plain and index millers, planers, trimming
presses, drill presses, sawmills, rock drills and mining machinery.
Later their mining and quarry-machinery business was moved to
Claremont, N. H., and became the Sullivan Machinery Company.
In 1889 the present Jones & Lamson Machine Company moved to
Springfield, Vt., where it now is. That same year, James Hartness
entered the employment of the company as superintendent. With his
advent the scattering of activities ceased and the Jones & Lamson
Machine Company began concentrating on turret lathes, which
Robbins & Lawrence and their various successors have been
manufacturing continuously since the early fifties. A number of the
old mechanics and foremen, who had homes in Windsor at the time
the company was moving to Springfield, took over the old shops and
organized the present Windsor Machine Company which now
manufactures the Gridley Automatic Lathes.
This, briefly, is the history of the old Robbins & Lawrence shop.
The men, however, who worked with Robbins & Lawrence and its
successors, are of greater interest.
While Lawrence continued as master-armorer of the Sharps Rifle
Works, the company was successful financially. Fitch, in the Census
article frequently referred to, says that he brought with him “from
Windsor the first plain milling machine used in Hartford.” Lawrence
also applied the broaching process to the manufacture of Sharps
rifles, effecting great economies, and was the inventor of the split
pulley which was first made for him at Lincoln’s Phœnix Iron Works.
In the winter of 1850 Lawrence introduced the practice of lubricating
rifle bullets with tallow, making possible the repeating rifle which had
been a failure up to that time as the barrel “leaded” and the gun lost
its accuracy. This was done in connection with some trials of the
Jennings rifle during the visit of Louis Kossuth, the Hungarian patriot,
who was in this country for the supposed purpose of purchasing
rifles.[186] Mr. Lawrence left the Sharps company in 1872 and was for