Professional Documents
Culture Documents
DSS Completed-1
DSS Completed-1
DSS Completed-1
BY
JANUARY, 2022
1
DECLARATION
written by me and is a record of my research work. To the best of my knowledge, it has not been
submitted or accepted in any institution of higher learning wholly or in part of any degree
elsewhere. All sources of information have been appropriately cited and referenced.
_____________________________ ________________________
17246044
2
CERTIFICATION
number 17246044, has been read and approved having met part of the
___________________________ ______________________
Ugheoke Iyenagbe Benjamin Date
(Supervisor)
_________________________ ______________________
(Head of Department)
_________________________ ______________________
3
DEDICATION
This project is dedicated foremost to God Almighty for His favor, mercy, and grace upon my life
throughout my academic years at the University of Abuja. His guidance and blessings have been
instrumental in my journey.
I would also like to dedicate this project to my parents, siblings, and friends for their unwavering
love, support, and encouragement. Their belief in me and constant encouragement have been a
I extend my heartfelt gratitude to everyone else who contributed to making my Bachelor's degree
pursuit a successful one. Your support, whether academic, emotional, or financial, has played a
I am truly grateful for the love, care, and assistance I have received from all these individuals.
This dedication is a testament to their profound impact on my life and academic journey
4
ACKNOWLEDGMENTS
I would like to express my heartfelt appreciation to God almighty for guiding me in choosing
Mechanical Engineering as my discipline and for providing me with the strength and
determination to complete this work successfully.
I am deeply grateful to my supervisor, Dr. Ugheoke Benjamin Iyenagbe, for his invaluable
support and mentorship throughout this research. His guidance and fatherly supervision have
been instrumental in shaping this project.
I extend my sincere thanks to Dr. Adiat Arogundade, the Head of the Department of Mechanical
Engineering, for her continuous support and encouragement throughout my academic journey.
I am indebted to the entire academic staff of the Department of Mechanical Engineering for their
commitment to providing effective and comprehensive academic work in field of study. Their
sacrifices and dedication have played a significant role in shaping my academic and personal
growth.
I would like to express my deepest gratitude to my parents, Mr. and Mrs. BELLO, for their
unwavering love, prayers, encouragement, motivation, and financial support throughout my
undergraduate years. Their belief in my abilities, has been a constant source of strength and
inspiration.
Lastly, I am grateful to all my friends and well-wishers who have supported me in various ways
during this academic endeavor. Your words of encouragement and belief in my abilities have
been invaluable.
5
TABLE OF CONTENT
Title page 1
Declaration 2
Certification 3
Dedication 4
Acknowledgement 5
Table of Content 6- 7
Abstract 8
INTRODUCTION
9
CHAPTER TWO
6
2.3.1 Architecture 18-19
2.4 Summary 19
CHAPTER THREE
CHAPTER FOUR
CHAPTER FIVE
REFERENCES 39-40
7
ABSTRACT
This study focuses on the development of a Decision Support System (DSS) for inventory
management in a pharmaceutical retail outlet. For the optimization of inventory management in
the pharmaceutical retail outlet to ensure the availability of essential medications based on
demand rate. in essences, predicting the reorder point of every drug available in the
pharmaceutical retail inventories database. Knowing the reorder point will in turn minimizing
costs and avoiding stockouts or excess inventory. The proposed DSS leverages the techniques of
reorder point analysis, to enhance decision-making in inventory management.
The DSS incorporates, the average demand-rate, lead time, and stock-level data, to predicting
the reorder point of each medicine. Then afterwards comparing the current stock-level with the
forecast reorder point and sends a feedback to the managerial team suggesting the current state
of the stock-in. this message will help to optimize its inventory levels and ordering decisions.
Furthermore, The system provides users with a user-friendly interface and graphical
representations, facilitating easy interpretation of inventory data and decision-making processes.
The development of this DSS aims to improve the effectiveness and efficiency of inventory
management in the pharmaceutical retail outlet. By utilizing forecast reorder point and
incorporating key inventory management parameters, the system enables proactive decision-
making, reducing the risk of stockouts or excess inventory. Ultimately, the implementation of the
DSS is expected to enhance overall inventory control practices in the pharmaceutical retail
industry, leading to improved customer satisfaction, cost savings, and optimized reorder point.
8
CHAPTER ONE
INTRODUCTION
1.1 Background of the study
Inventory Management in a middleman retail industry, holds different types of merchandise and
quantity in stock, in order to meet the demand of customers all around every goods.
Demand rate varies from item to item and stocking of items to make up inventories is capital
intensive such that it is difficult to tie down money on goods that are not necessarily in demand.
In other words, having too much stock on hand ties up capital, while on the other hand, not
having the right quantity or enough merchandise in stock will often lead to loss of sales as well
company. It is therefore necessary to have a balance between carrying much quantity of an item
in inventory higher than the demand rate which would lead to capital being tied down or
allowing lean holding of the right quantity, below the demand rate, which could lead to stock-out
and its attendant consequences. Striking this balance would help to minimize losses and
maximize profit.
This challenge particularly affects retail pharmacies, where factors that affect sales of medicines
vary so rapidly. In such environments, the use of the classical inventory control or management
models is minimally helpful and sometimes not useful. In such an enterprise, it is known that
medicines could be very expensive, have divergent demand rates, with different lead times, and
hence reorder points, etcetera. It was thought that through the use of data science combined with
deep learning of historical records of sales and orders of established pharmacies, a better
inventory management system could be developed that would help to strike the balance between
9
out-of-stock and overstocking of medicine at any given time, based on the demand of rate of
each of the medicines in the holding of the pharmacy, which will result to maximizing profits,
satisfying all customers and therefore minimizing loss. Such a system would be a variant of a
decision support system since striking such a balance relies heavily on decision-making.
Decision Support System (DSS) takes data as input then uses data analysis tools to adjust itself to
forecast and respond based on what it knows from a series of data in the database history and
then suggests decisions to the management. DSS is a computer-based information system that
pharmaceutical retail outlet to make forecasts based on a demand rate. DSS is an interactive
information from a combination of data models to identify limitations and suggest possible
DSSs take data available in the database to generate desired information for the management, it
performs a logical test or a conditional statement to forecast accurate predictions necessary for
making an effective decision in the inventory management application. DSS requires a database,
which is an organized collection of data for one or more purposes. The data available in the
database are structured to model or schema in a way that supports the processing of decisions in
The term database implies that the data is managed to some level of quality (measured in terms
of accuracy, availability, usability, and resilience) and this in turn often implies the use of a
10
As retail is evolving at an accelerated rate, enterprises are increasingly focusing on how to use
artificial intelligence technology to reshape the retail industry’s ecology and integrate online and
offline experiences. Acquiring infinite access to goods in stock is limited due to the fact that the
demand for a particular medicine varies from another. However, in order to satisfy the customers
at all times within limited resources, there should be contrived ways to establish demand rates for
medicines and their order lead times. If these quantities are established, it is possible to develop a
Decision Support System (DSS) based on deep learning techniques, that would forecast and give
advice on the re-order point of any goods and ease other decision-making processes of the
This project aims to improve the efficiency of and maximize profit in pharmaceutical retail
outlets through the development of an optimized intelligent decision support system for
inventory management using Convolutional Neural Network (CNN) for scanning goods.
2. Development of a database.
segmentation.
11
Inventory Management Applications play a vital role in manufacturing systems. Deploying the
same to a pharmaceutical retail outlet would help boost efficiency and improve profits. This
would be because the reorder points, lead times, reorder quantity, etcetera, of the different
medical holdings of the pharmacy which would have been managed manually in a cumbersome
manner, would be processed seamlessly and in real-time in the DSS application. Thus, a decision
support system in an inventory management application would that the right amount of goods
and the type of goods will always be available in stock. This means there will be the availability
improved customer satisfaction experience. This would lead to higher sales at lower human
mental tasks and overall better performance and management of pharmaceutical or patent
medicine shops.
The study is limited to the development of a full-stack DSS application. The dataset used set for
training the model or algorithm for the recognition of medicine through image segmentation
using VGG-16 Architecture is based on another project. The application can be used mostly in a
pharmaceutical retail outlet to perform the identification of drugs by scanning using CNN.
12
CHAPTER TWO
LITERATURE REVIEW
2.1 Overview Of DSS in Inventory Management
Several studies have explored the use of DSS for inventory management based on the reorder
point, Inventory management is a critical aspect of the pharmaceutical retail industry, where
products have a limited shelf life and need to be handled with care. The reorder point is an
Decision Support Systems are computer-based systems that help decision-makers to analyze and
solve complex problems. In inventory management, DSS can provide insights into demand
forecasting, inventory optimization, and supply chain management. DSS can also help to reduce
stock-outs, overstocking, and waste, thereby improving profitability and customer satisfaction.
According to Keen (1978), The technical work on interactive computer systems was mainly
carried out at the Massachusetts Institute of Technology in the 1960s. It is considered that the
concept of DSS became an area of research of its own in the middle of the 1970s, before gaining
in intensity during the 1980s. In the 1970s DSS was described as “a computer-based system to
aid decision making”. In the late 1970s, the DSS movement started focusing on “interactive
computer-based systems which help decision-makers utilize databases and models to solve ill-
13
structured problems”. In the 1980s DSS should provide systems “using suitable and available
technology to improve the effectiveness of managerial and professional activities” [1], [2].
According to Uribe Cadavid (2011), The statement on forecast discusses the use of time series
models for forecasting in inventory management and the different models that can be used, such
as simple moving averages and exponential smoothing. It also highlights the importance of using
forecast demand and standard deviation of forecast errors instead of average demand and
standard deviation in inventory control models to prevent excessive inventory or backlogs. The
objective of developing a decision support system for inventory management is to use historical
demand data and time series models to improve the accuracy of forecasting and aid in making
Across various research papers, the definition of a decision support system has been reviewed
basically as systems that help in managerial precision through decision-making, DSS is a system
of specific information intended to assist the management in taking decisions relating to issues
that are semi-structured, and does not replace the function of the decision maker to make
decisions [3].
Decision Support Systems (DSS) can also be defined as interactive computer-based systems
intended to help decision-makers utilize data and models in order to identify and solve problems
According to Mora et al., the decision maker employs computer technology to [5], [6]:
14
(a)organize the information into problem factors,
Defining standard characteristics of DSS can be summarized by Turban and Aronson as follows
[7]:
The major features that distinguish DSS (Decision Support Systems) from other previously
established systems. These features are identified by Turban and Aronson and include covering a
vast spectrum of managerial levels, providing support for both individuals and groups,
facilitating interdependent and sequential decisions, carrying out all parts of the decision-making
process, covering a variety of decision analysis tools, being adaptive and flexible, having strong
rather than replacing them. The summary highlights the user-friendly and control elements of
DSS as well as its focus on appropriateness and quality in decision-making rather than just
efficiency.
2.1.3 Decision-Making
According to Lemass, DSS (Decision Support System) should aim to improve both the
15
Effectiveness is defined as the degree to which identified goals are achieved, while efficiency
measures the application of resources to attain those goals. Decision-making involves assessing
consequences, making the choice based on logical judgment, informing others, and evaluating
decisions. A DSS is a system that can ease this process, particularly in predicting stock levels to
avoid excess or shortages. The statement also highlights the aim of developing a DSS for
re-order points based on demand rate during the lead time and safety stock using a model or
algorithm [9].
selection of one course of action among multiple strategies and DSS is a prevailing system that
can ease this process [10], which raised the problem of how to predict the stock of each product
to the right in order to avoid excess/shortages. Therefore, this study aims to develop a decision
The DSS in a pharmaceutical retail outlet will facilitate the decision-making of re-order points
based on the demand rate during the lead time and the safety stock using a model or algorithm.
DSS for some other application areas with an emphasis on DSS. Application Areas such as;
Medical, Management, Manufacturing industry, Stock trading, and Weather forecasting [12]
The article reviews the use of decision support systems (DSS) in various application areas such
and mathematical models, data analysis, and variable relationships. The importance of decision-
making in managing organizations is highlighted, and DSS is seen as a valuable tool for
control [13]. Inventory management is a function that is responsible for all decisions regarding
the stock in an organization. In inventory management made policies, activities and procedures
to ensure the proper amount of stock of each product for a certain time [14].
In addition, most studies of inventory management are associated with Supply Chain
management in the pharmacy are still very few [15], Therefore, it is developing a decision
support system (DSS) for pharmacy inventory management is a way to predict the next period of
purchase of pharmaceutical products with other approaches apart from time series, in order to
reduce excess stock and avoid shortages, so the costs associated with inventory management in
A huge amount of work has been done by researchers to improve the CNN architecture on
features or patterns for deep learning convolutional neural networks, therefore knowledge has
been gained over the years which led to the development of various models for information
17
retrieval. These models have brought about significant improvements in all aspects of deep
learning [16].
CNN’s algorithm is analogous to the traditional Artificial Neural Networks (ANNs) in that, they
are comprised of neurons also called nodes that self-optimize through learning by extracting
features or patterns directly from the data set. CNN can also be said to be computer vision, which
and ML are techniques that have been around for quite a while [20]. They are specifically
focused on teaching an algorithm how to learn from a data set and produce some information but
can’t explicitly extract features or patterns directly from the data set. DL is focused on the use of
neural networks, it uses the concept of a neural network that is derived from the structure of the
At present, barcode recognition is the most widely used technology not only in research but also
on each product package, the management of products can be easily facilitated. Normally, almost
every item on the market has its corresponding barcode. However, due to the uncertainty of the
printing position of the barcode or if a barcode is damaged, there would be no way to scan the
product. it often requires time to manually find the barcode and assist the machine in identifying
the barcode at the retail outlet. Based on a survey from Digimarc, 45% of customers complained
that, sometimes, it was not convenient to use barcode scanning machines [18].
2.3.1 Architecture
[19]. The final output has two channels as score maps: the region score and the affinity
18
score[19]. with batch, normalization is adopted as our backbone. This is the interface model that
was used in Youngmin’s paper in order to obtain text identification and recognition.
Our model must skip connections in the decoding part, which is similar to U-net [20] in that it
2.4 Summary
In conclusion, DSS has been applied in many areas of the industry such as the medical field,
agricultural field, Management, and Manufacturing industry, etc. however from the extent of
literature the application of DSS in inventory control and management in pharmaceutical retail
shops is not reported and this is the gap this project work is intending to fill. Decision Support
19
Systems can provide valuable insights and recommendations for setting the reorder point in
pharmaceutical retail outlets. The literature review highlights the potential of DSS based on
optimizing the reorder point. Implementing a DSS can help retailers to reduce costs, improve
CHAPTER THREE
Materials and Methods
Back-End Front-End
Application- GUI
Server (Pages)
Login/Dashboard Login/Dashboard
Add/remove
product Invoice generator
Print report
Figure 2: Overview of the Application
The development of a Decision Support System (DSS) for Inventory management in a
pharmaceutical retail outlet requires a systematic methodology. Below are the steps involved in
1. Define the problem: The first step in developing a DSS is to identify the problem. In this
and overstocking.
2. Gather data: The next step is to collect data on the current inventory levels, sales, and
customer demand. The data will be used to develop a model for inventory management.
3. Develop the model: Based on the collected data, a model for inventory management is
developed. The model will take into account factors such as lead time, safety stock,
4. Implement the DSS: The model will be integrated into a DSS, which will provide
decision support to the retail outlet. The DSS will generate recommendations for
5. Test and validate: The DSS will be tested and validated using historical data. The
performance of the DSS will be evaluated based on its ability to reduce stockouts and
21
6. Monitor and update: The DSS should be monitored regularly to ensure that it is
performing optimally. The DSS should be updated as necessary based on changes in sales
Below is a flowchart that describes the working process of DSS in the application including DSS
22
Figure 3: Flow chart
23
1. Item characteristics
4. Demand variability
5. Order Quantity
Are required for the explicit development of the Application, since the application is software
based it requires a few items for the development. Once a drug is identified through scanning, it
then goes through the item list stored in the database and then outputs the details of the particular
drug which was scanned, details such as quantity available and other relevant info will be
displayed.
The DSS according to the model will facilitate re-order points for each pharmaceutical drug
available in the store and further promote profitability. DSS is dedicated to predicting re-order
points Since such a prediction will lead to effective customer service and improve the efficiency
To obtain the Reorder point there are some essential variables that should be available such as:
1. Stock in
2. Demand rate
3. Lead time
24
If the variables above are known the reorder point can be predicted, below are procedures for
Firstly we will obtain the safety stock for a particular drug, if the demand rate has been obtained
The demand rate shows the rate of purchase of a particular drug leaving the stock daily.
The demand rate, or demand, refers to the quantity of a product or service that customers are
willing to purchase during a given period. It represents the rate at which customers consume or
purchase a product or service. The demand rate can be expressed as the number of units sold or
1. Historical sales data: A company can analyze past sales data to determine the demand rate
for a particular product or service. This involves looking at the number of units sold or
Once the demand rate has been obtained, it can use this information to make strategic
estimating the demand rate, can optimize its operations and increase profitability.
Then lead time, which is the time taken between the placement of an order for stocks to the time
25
Lead time is the amount of time between placing an order for a product or material and receiving
it. It represents the time required for the supplier to process the order, manufacture or acquire the
product, and transport it to the buyer. Lead time can include both processing time and
transportation time.
1. Historical data: A company can analyze past orders to determine the lead time for a
particular product or material. This involves tracking the time between placing an order
By accurately estimating lead time, we can then improve its inventory management, reduce the
risk of stockouts or overstocking, and optimize its production and supply chain operations
To obtain the re-order point for a particular medicine, we did an analysis by:
Predicting the reorder point is an important task in inventory management that involves
determining the optimal time to place an order for a product to avoid stockouts and maintain
adequate inventory levels. Here are some steps you can follow to predict the reorder point:
1. Determine the lead time: The lead time is the time it takes for the supplier to deliver the
ordered products. It is important to determine this time accurately to know when to place
the order.
2. Calculate the demand during the lead time: You need to estimate the demand for the
product during the lead time. This can be done using historical sales data or by using
forecasting techniques.
26
3. Determine the safety stock: The safety stock is the extra inventory you keep to cover
4. Calculate the reorder point: The reorder point is the inventory level at which you need to
place an order to avoid a stockout. It can be calculated using the following formula:
Once you have determined the reorder point, then we can set up a system to trigger an order
when the inventory level reaches that point. It is important to regularly review and adjust the
reorder point based on changes in demand, lead time, and safety stock levels.
lead_time = 7 # days
safety_stock = 50 # units
The result
27
After every purchase, the program should display a pop-up message indicating the current status
of a certain drug. If the reorder point has been reached, a message for re-order should be
The program should compare each purchase with the reorder point forecast to confirm whether
the inventory has reached the reorder point. If the inventory equals or is less than the reorder
point, a message should be triggered to inform management that a particular medicine has
reached reorder point or is out of stock. This message should include a specific output with a
The Database of this application is where the data of this application will be stored and will also
be accessible at any time by the managerial team. The structured query language (SQL), will be
used to obtain the creation of data, reading of data, update of data, and including deleting of data.
These operations are the common operation that can be performed by every database, and in the
development of this project work, these objectives will be possible in order to add, remove or
To use the mongo.py library for a database in a pharmaceutical retail outlet application, you can
1. Install the pymongo library: First, you need to install the pymongo library. This can be
done by running the following command in your terminal: pip install pymongo.
2. Set up a MongoDB database: Next, you need to set up a MongoDB database. You can
either set up a local MongoDB database or use a cloud-based MongoDB service such as
MongoDB Atlas.
28
3. Connect to the database: Once you have set up your MongoDB database, you can connect
to it using the pymongo library. You will need to specify the connection URL and any
required credentials. Here is an example code snippet for connecting to a local MongoDB
database:
client = MongoClient('mongodb://localhost:27017/')
db = client['my_database']
4. Create collections and documents: Once you have connected to the database, you can
create collections and documents to store your data. Here is an example code snippet for
collection = db['products']
document = {
'name': 'Paracetamol',
'price': 5.99,
'stock': 100,
collection.insert_one(document)
5. Query the database: You can query the database to retrieve data using the find() method.
Here is an example code snippet for querying the products collection and printing the
results.
collection = db['products']
29
results = collection.find({})
print(result)
These are the basic steps for using the mongo.py library for a database in a pharmaceutical retail
outlet application.
To set up Python as the server for a pharmaceutical retail outlet application, we can follow these
steps:
1. Install Python: If you haven't already, you need to install Python on your computer. You
can download the latest version of Python from the official website at python.org.
2. Install a package manager: A package manager is a tool that makes it easy to install and
manage software packages, including Python libraries. There are several package
managers available for Python, including pip and Anaconda. For a pharmaceutical retail
3. Install required libraries: Once you have pip installed, you can use it to install any Python
libraries that your application requires. Some common libraries for a pharmaceutical
retail outlet application might include Flask (for building a web application), pymongo
(for working with a MongoDB database), and pandas (for data analysis and
manipulation).
4. Write your application: Once you have your development environment set up, you can
start writing your application. You can use any text editor or integrated development
30
environment (IDE) to write Python code. Some popular choices include Visual Studio
5. Test and deploy your application: Once you have written your application, you can test it
to make sure it works as expected. You can use Python's built-in testing framework (unit
test) or a third-party testing library such as pytest. Once you are satisfied with your
application, you can deploy it to a web server or cloud service for production use.
These are the basic steps for setting up Python for a pharmaceutical retail outlet application. The
specific requirements and details of your application may vary depending on the specific needs
of your business.
To set up HTML, Bootstrap CSS, and JavaScript for a pharmaceutical retail outlet application,
1. Choose an HTML editor: HTML can be written using any text editor, but using an HTML
editor with features such as syntax highlighting and autocompletion can make the process
much easier. Popular HTML editors include Visual Studio Code, Atom, and Sublime
Text.
2. Include the Bootstrap CSS and JavaScript files: Bootstrap is a popular CSS framework
that can be used to create responsive and visually appealing websites. You can download
the Bootstrap CSS and JavaScript files from the Bootstrap website and include them in
your HTML file. You can also link to the files directly from a Content Delivery Network
31
3. Create a basic HTML file: Start by creating a new file and saving it with the ".html"
extension. In the file, start with the basic HTML structure by adding the HTML, head,
and body tags. Inside the body tag, add some content to get started.
4. Add Bootstrap classes to your HTML elements: Bootstrap includes a set of predefined
classes that can be added to HTML elements to apply prebuilt styles. You can add classes
such as "container", "row", and "col" to create responsive grids, or "btn" to create buttons
5. Add custom CSS: To further customize your application, you can add your own CSS
rules. You can create a separate CSS file or add your styles directly in the HTML file
interactivity to your HTML pages. You can use JavaScript to create dynamic forms,
7. Test your application: Once you have created your HTML, CSS, and JavaScript files, you
can test your application by opening the HTML file in a web browser. Make sure all the
elements are displaying correctly and that any JavaScript functions are working as
intended.
These are the basic steps for setting up HTML, Bootstrap CSS, and JavaScript for a
Convolutional Neural Network (CNN) will be used to implement the scanning of the medicine
32
Vision to a computer is a very complex task to be done, compared to a human eye that can
perform the task with ease, CNN resembles the human way of visual recognition. It receives
pixels as image inputs, transforms them using mathematical computations and activation
functions then builds a classifier for the input classes. CNN has an encode-specific feature in its
architecture that suits it to perform image processing and computer vision tasks.
The computer visualizes images as a group of pixels that has a degree of weight attached to
every set of pixels, using a paper written by Youngmin, the architecture described there would be
used for the identification of pharmaceutical drugs. The intention of CNN is to extract features
extracting a set of text from the package of the drug through word segmentation on every image
captured, an algorithm at the fully connected layer of the CNN for text detection and recognition
This will facilitate the management of pharmaceutical products and improve consumer shopping
experience. The main objective is to enable the identification of each character of words in
natural images.
features from the pack of the drug once the drug is identified through comparing with its dataset
used in training the model for drug recognition and identification. Afterward, it outputs the
details of the drug at the point of sale. Using a Convolutional Neural Network (CNN) to scan a
33
1. Import necessary libraries: The first step is to import the necessary libraries such as
2. Collect data: The first step is to collect a large dataset of product images. This dataset
should include images of all the products that the CNN will be required to scan.
3. Preprocess the data: The dataset needs to be preprocessed before being used to train the
CNN. This includes tasks such as resizing images to a standard size and normalizing the
pixel values.
4. Define the CNN architecture: The next step is to define the architecture of the CNN. This
includes specifying the number and type of layers in the network, such as convolutional
5. Compile the model: Once the architecture is defined, the model needs to be compiled
6. Train the model: The next step is to train the model using the prepared dataset. This
involves fitting the model to the training data and adjusting the weights of the layers to
7. Evaluate the model: Once the model is trained, it needs to be evaluated on the testing
dataset to measure its performance and identify any areas for improvement.
8. Save the model: Once the model is trained and evaluated, it can be saved for later use in
9. Integrate the model: Finally, the model can be integrated into the scanning system of the
the products.
34
10. Monitor and update the CNN: Finally, it is important to monitor the performance of the CNN
over time and update it as needed. This includes retraining the CNN with new data, adjusting the
CNN's parameters, and improving the dataset to include more diverse product images.
Programming a CNN for scanning products can be a complex task, and it requires knowledge of
programming, machine learning, and computer vision. By using a CNN to scan products in a
pharmaceutical retail shop as a tool for image segmentation, it is possible to automate the process
of product identification and reduce errors that may occur during manual scanning. However, it
is important to ensure that the CNN is trained and tested appropriately to ensure the accurate
classification of products.
3.6 Summary
Above are the methods for creating a DSS for Inventory Management in a pharmaceutical retail
outlet, The coding of the application involves each of those categories listed previously in
(figure2). in other to have a deployed application of the DSS in inventory management, and
therefore with frequent updates of the application the software can indeed be marketable.
35
CHAPTER FOUR
RESULTS AND DISCUSSION
4.1 DSS Program Result
The implementation of the DSS for inventory management in the pharmaceutical retail outlet
1. Reduced Stockouts
2. Increased Profitability
5. Cost savings
with the application, having an employee page and also an admin page, the employee
handles the purchase section while the managerial side handles the entire management of
the pharmaceutical retail store. It provides real-time visibility into inventory levels,
2. Enhanced decision-making with DSS: By utilizing accurate data, the DSS assists in
reorder points, safety stock levels, and order quantities based on demand rate, leading to
36
more precise and optimal inventory decisions. This leads to a reduction in stockouts,
overstocking, and excess inventory, which in turn improves overall operational efficiency.
But to yet be integrated fully into the inventory software to interact with the datebase.
3. The database: the database can perform the CRUD function that is to create, read, update,
4. CNN for scanning: the CNN algorithm is currently in the learning stage, learning
different medicine product names, product categories, and sub-categories for the
Overall, implementing a DSS for inventory management in a pharmaceutical retail outlet can
lead to optimized inventory levels, improved decision-making, reduced costs, enhanced customer
A Decision Support System (DSS) for inventory management in a pharmaceutical retail outlet
typically consists of several key components. Here are some essential components:
1. Data Management
2. Forecasting reorder_point
3. Inventory Optimization
37
These components work together to create a comprehensive DSS for inventory management in a
pharmaceutical retail outlet. The integration of these components facilitates effective decision-
The economics of acceptability in a Decision Support System (DSS) for inventory management
in a pharmaceutical retail outlet refers to evaluating the cost-effectiveness and financial viability
of implementing and using the DSS. Here are some key considerations related to the economics
of acceptability:
1. Cost of Implementation
2. Operational Costs
4. Long-Term Sustainability
By analyzing the economics of acceptability, you can determine whether implementing a DSS
for inventory management in a pharmaceutical retail outlet is financially viable and aligns with
the organization's objectives. It enables you to make informed decisions about investing in the
DSS and ensures that the economic benefits outweigh the associated costs.
38
CHAPTER FIVE
5.0 Conclusion and Recommendation
5.1 Conclusion
The implementation of a DSS for inventory management in a pharmaceutical retail outlet can
result in significant benefits, including optimized inventory levels, reduced stockouts, increased
profitability, and automated inventory management processes. The success of the DSS depends
on the quality of the data used to develop the model, as well as the accuracy of the
recommendations generated by the DSS. With proper training and support, the DSS can become
5.2 Recommendation
1. CNN: Explore CNN algorithms to determine computer vision for scanning product is a
very complicated program, furthermore thus the program would require a more
2. Collect and analyze relevant data: Gather accurate and up-to-date data on sales, inventory
levels, supplier lead times, and customer demand. Analyze this data to identify patterns,
trends, and seasonality that can inform your inventory management decisions.
implementing a DSS for inventory management strategies to optimize efficiency, reduce costs,
39
REFERENCE
[1] Rajni Jain and S.S Raju (2021). Decision Support System in Agriculture using Quantitative
Analysis
[2] Keen, P. G. W. (1978). Decision support systems: an organizational perspective. Reading,
Mass., Addison-Wesley Pub. Co. ISBN 0-201- 03667-3
[3] Uribe Cadavid, D.C. & Castro Zuluaga, C. (2011). A framework for decision support system
in inventory management area. Paper presented at the 9th Latin American and Caribbean
Conference for Engineering and Technology, August 3-5, 2011, Medellín, Colombia. SOCODA,
Medellín, Colombia, diana.uribe@socoda.com.co and Universidad EAFIT, Medellín, Colombia.
[4] Power, D. J. (1999). Decision Support Systems Glossary. Retrieved from
http://DSSResources.COM/glossary/
[5] Mora, M., Forgionne, G., & Gupta, J. (2003). Decision Making Support Systems:
Achievements and Challenges for the New Decade. Natural Resources Planning, Silva Fennica:
Harrisburg, PA.
[6] Decision Support Systems Maria Rashidi, Maryam Ghodrat, Bijan Samali and Masoud
Mohammadi
[7] Turban, E., & Aronson, J. (2001). Decision Support Systems and Intelligent Systems. New
Delhi, India: Prentice Hall.
[8] Lemass, B. (2004). Structured Conceptual Design: The New Frontier. Sydney: Pearson
Prentice Hall. [2]
[9] Bohanec, M. (2001). What is Decision Support? [jozef Stefan institute]. Retrieved [2001.
[10] Riedl, F., & Holsapple, C. (2008). Handbook on Decision Support Systems. New York:
Springer-Verlag.
[11] Gustriansyah, R., Sensuse, D., & Ramadhan, A. (2015). Decision Support System for
Inventory Management in Pharmacy Using Fuzzy Analytic Hierarchy Process and Sequential
Pattern Analysis Approach.
[12] Development of a Decision Support System for Inventory Analysis and Control Numera
Tahir and Muhammad Abbas Choudhary
[13] McKay, K. C. (2008). Transforming Inventory into a Competitive Advantage: Best Practices
and Productivity Tools. Defense Transportation Journal, 64(2), 123-125.
[14] Waters, D. (2003). Inventory Control and Management (2nd ed.). England: John Wiley &
Sons Ltd. (pp. 3-4).
[15] Kelle, P., Woosley, J., & Schneider, H. (2012). Pharmaceutical supply chain specifics and
inventory solutions for a hospital case. Operations Research for Health Care, 1(2-3), 54-63.
40
[16] Adewumi, E. A. (2020). A Convolutional Neural Network (CNN) based Pill Image Retrieval
System.
[17] Owino, V. O. (2020). Deep learning Convolutional Neural Network Image Classification (Python):
Computational neuroscience project [Unpublished manuscript]
[18] Wei, Y., Tran, S., Xu, S., Kang, B., & Springer, M. (2020). Deep Learning for Retail Product
Recognition: Challenges and Techniques.
[19] Baek, Y., Lee, B., Han, D., Yun, S., & Lee, H. (2019). Character Region Awareness for Text
Detection.
[20] Ronneberger, O., Fischer, P., & Brox, T. (2015). U-net: Convolutional networks for
biomedical image segmentation. In MICCAI (pp. 234-241). Springer.
41