PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS

AVAILABILITY IN GOVERNMENT HOSPITALS

A PROJECT REPORT

Submitted by,
MADHU KUMAR V 20211CSE0363
RAVI SHIVAJI MAHIPATI 20211CSE0374
TARUN GS 20211CSE0385

Under the guidance of,

Ms. Shweta Singh
Assistant Professor
School of Computer Science and Engineering,
Presidency University, Bengaluru

in partial fulfillment for the award of the

degree of

BACHELOR OF TECHNOLOGY

IN

COMPUTER SCIENCE AND ENGINEERING

At

PRESIDENCY UNIVERSITY
BENGALURU
DECEMBER 2024
 PRESIDENCY UNIVERSITY

SCHOOL OF COMPUTER SCIENCE ENGINEERING

CERTIFICATE
This is to certify that the Project report “PREDICTIVE ANALYSIS ON
MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT
HOSPITALS” being submitted by “Madhu Kumar V , Ravi Shivaji Mahipati,
Tarun GS ”bearing roll numbers “20211CSE0363,20211CSE0374,20211CSE0385”
in partial fulfillment of the requirement for the award of the degree of Bachelor of
Technology in Computer Science and Engineering is a bonafide work carried out
under my supervision.

MS.SHWETA SINGH DR . ASIF MOHAMMED

Assistant Professor Assistant Professor & HoD
School of CSE&IS School of CSE&IS
Presidency University Presidency University

Dr. L. SHAKKEERA Dr. MYDHILI NAIR Dr. SAMEERUDDIN KHAN

Associate Dean Associate Dean Pro-VC School of Engineering
School of CSE School of CSE Dean -School of CSE&IS
Presidency University Presidency University Presidency University
 PRESIDENCY UNIVERSITY

SCHOOL OF COMPUTER SCIENCE ENGINEERING

DECLARATION

We hereby declare that the work, which is being presented in the project report
entitled PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS
AVAILABILITY IN GOVERNMENT HOSPITALS in partial fulfillment for
the award of Degree of Bachelor of Technology in Computer Science and
Engineering, is a record of our own investigations carried under the guidance of
MS.SHWETA SINGH, Assistant Professor, School of Computer Science
Engineering, Presidency University, Bengaluru.

We have not submitted the matter presented in this report anywhere for the award of
any other Degree.

NAME ROLL NO SIGNATURE

MADHU KUMAR V 20211CSE0363
RAVI SHIVAJI MAHIPATI 20211CSE0374
TARUN GS 20211CSE0385
 ABSTRACT
Government hospitals are lifelines for millions of people in India, especially for those
who rely on public health services. They provide essential treatment, often free of cost,
ensuring that healthcare is accessible to all. However, during critical times such as disease
outbreaks, these hospitals often face overwhelming challenges. A recurring issue is the
shortage of essential medicines and doctors, which can lead to long waiting times, unmet
patient needs, and compromised quality of care. This project focuses on harnessing the
power of predictive analytics to address these issues, ensuring that hospitals are prepared
to meet patient demand effectively and efficiently.

Medicine shortages during peak disease periods are a common problem in government
hospitals. Patients often find that vital medicines are unavailable when they need them
most. These shortages arise from several factors, including unanticipated surges in
demand, inadequate forecasting, and supply chain delays. For instance, during the
monsoon season, diseases like dengue and malaria spike, and the demand for specific
medicines increases dramatically. Without proper planning, hospitals struggle to cope.
This project uses historical patient data, current disease trends, and regional patterns to
predict the types and quantities of medicines required at specific hospitals during
particular times. By generating detailed reports, the system enables hospital
administrators to stock the right medicines in the right quantities at the right time,
reducing the risk of shortages. Patients receive the treatments they need without
unnecessary delays, enhancing trust in the public healthcare system.

Equally critical is the availability of doctors and specialists, particularly during disease
outbreaks or times of increased patient inflow, such as weekends, holidays, or evenings.
Patients often experience frustration when they cannot find the required doctor, especially
during emergencies. For example, during the COVID-19 pandemic, many hospitals faced
challenges in managing the influx of patients needing critical care. Predictive analytics
can help address this problem by analyzing historical patient flow data, current disease
patterns, and anticipated trends. The system forecasts the number of doctors required in
different departments and generates actionable insights for hospital administrators. This
enables better workforce planning, ensuring that hospitals have adequate medical staff to
meet patient needs, even during peak periods. Furthermore, the system can identify the
need for specialists based on the nature of diseases prevalent at a particular time, helping
hospitals deploy their resources more effectively.

This initiative is tailored to meet the needs of the Indian Government healthcare system,
focusing on using analytics to improve operational efficiency. By leveraging data-driven
insights, the program seeks to eliminate inefficiencies, minimize resource wastage, and
ensure that hospitals are better equipped to serve their patients. The system provides
actionable intelligence that allows administrators to make informed decisions, anticipate
future challenges, and allocate resources proactively. This not only improves the day-to-
day functioning of hospitals but also strengthens the overall healthcare infrastructure,
making it more resilient and capable of handling crises.
 ACKNOWLEDGEMENT
First of all, we indebted to the GOD ALMIGHTY for giving me an opportunity to
excel in our efforts to complete this project on time.
We express our sincere thanks to our respected dean Dr. Md. Sameeruddin Khan,
Pro-VC, School of Engineering and Dean, School of Computer Science Engineering &
Information Science, Presidency University for getting us permission to undergo the
project.
We express our heartfelt gratitude to our beloved Associate Deans Dr. Shakkeera L
and Dr. Mydhili Nair, School of Computer Science Engineering & Information
Science, Presidency University, and “Dr Asif Mohammed” Head of the Department,
School of Computer Science Engineering & Information Science, Presidency
University, for rendering timely help in completing this project successfully.
We are greatly indebted to our guide Ms .Shweta Singh, Assistant Professor and
Reviewer Ms.Ranjitha Professor, School of Computer Science Engineering,
Presidency University for his inspirational guidance, and valuable suggestions and for
providing us a chance to express our technical capabilities in every respect for the
completion of the project work.
We would like to convey our gratitude and heartfelt thanks to the PIP2001 Capstone
Project Coordinators Dr. Sampath A K, Dr. Abdul Khadar A and Mr. Md Ziaur
Rahman and Git hub coordinator Mr. Muthuraju V.
We thank our family and friends for the strong support and inspiration they have
provided us in bringing out this project.

Madhu kumar v
Ravi Shivaji Mahipati
Tarun GS
LIST OF FIGURES

Sl. No. Figure Name Caption Page No.

1 Fig 6.1 flow chart of doctor 23
and medicine
prediction

ii
 TABLE OF CONTENTS

CHAPTERNO. TITLE PAGENO.

ABSTRACT
ACKNOWLEDGMENT
1. INTRODUCTION 1
2. LITERATURE REVIEW 5
3. RESEARCH GAPS OF EXISTING METHODS 12
4. PROPOSED METHODOLOGY 14
5. OBJECTIVES 18
6. SYSTEM DESIGN & IMPLEMENTATION 22
7. TIMELINE FOR EXECUTION OF PROJECT 24
8. OUTCOMES 25
9. RESULTS AND DISCUSSIONS 28
10. CONCLUSION 30
11. REFERENCES 32

iii
 PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT HOSPITALS

CHAPTER-1
INTRODUCTION
Government hospitals are the heart of healthcare in India, offering essential medical services
to millions, particularly those who cannot afford private healthcare. These institutions stand
as a lifeline for many, yet they often grapple with challenges, especially during peak disease
seasons. Medicine shortages and the unavailability of doctors during critical times can lead to
delayed treatments, leaving patients frustrated and vulnerable. Recognizing these pressing
issues, this project proposes a predictive analytics system that empowers hospitals to
anticipate and manage resource demands. By leveraging data, the system aims to ensure timely
care for patients, alleviating stress on healthcare providers and improving overall operational
efficiency.

1.1 Why Predictive Analytics is Essential for Government Hospitals

Government hospitals operate under tight constraints—limited budgets, unpredictable patient
inflows, and sudden disease outbreaks. These challenges make it difficult to plan resources
effectively. Predictive analytics can turn the tide by enabling hospitals to proactively prepare
for surges in demand, making healthcare more responsive and efficient.

1.1.1 Tackling Medicine Shortages

Imagine visiting a hospital during a dengue outbreak, only to find that essential medicines like
platelet concentrates are unavailable. This is not just an inconvenience; it can be life-
threatening for patients. Medicine shortages during disease peaks are a common occurrence
in government hospitals.

Predictive analytics offers a solution by analyzing patient histories, seasonal patterns, and real-
time data to forecast which medicines will be needed, in what quantities, and where. For
example, during monsoon seasons, when waterborne diseases like cholera and typhoid spike,
the system can predict higher demand for antibiotics and rehydration solutions. This allows
hospitals to stock these medicines well in advance, avoiding the distress caused by shortages.

By ensuring that the right medicines are available at the right time, the system not only

School of Computer Science Engineering & Information Science, Presidency University.

2
 PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT HOSPITALS

improves patient outcomes but also builds trust in public healthcare. Moreover, it helps reduce
wastage by avoiding overstocking of less-needed medicines, striking a balance between
availability and cost efficiency.

1.1.2 Ensuring Doctor Availability

Another frequent complaint from patients is the unavailability of doctors when they need them
most—weekends, holidays, or during evenings. Patients arriving with urgent concerns often
face long waits or are unable to consult specialists. This issue is especially pronounced during
periods of high patient inflow, such as flu seasons or disease outbreaks.

The predictive analytics system tackles this problem by studying past and current data to
understand patient inflow patterns. For instance, if a hospital sees a trend of increased
respiratory issues during winter evenings, the system can recommend scheduling additional
pulmonologists during those hours. Similarly, for rural hospitals that face a rush on weekends,
the system can advise deploying more general physicians or specialists to handle the spike.

This proactive approach not only ensures that patients receive timely care but also helps
prevent doctor burnout. By aligning staffing levels with patient needs, the system creates a
balanced workload for healthcare providers, leading to better care quality and happier patients.

1.2 Who Benefits from This System and What It Aims to Achieve
1.2.1 Targeting the Right Stakeholders
This predictive analytics system is designed for the Indian Government healthcare department,
which oversees public hospitals nationwide. By implementing this system, the department can
address long-standing challenges in resource management and healthcare delivery.

Hospital administrators will also find immense value in the system. It equips them with
actionable insights to plan resources, anticipate demand, and respond to patient needs more
effectively. Patients, too, stand to benefit directly. With better medicine availability and timely
access to doctors, their experience at public hospitals will improve significantly.

1.2.2 Driving Key Improvements

The project aims to drive meaningful change in the way government hospitals operate. The
School of Computer Science Engineering & Information Science, Presidency University.
3
 PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT HOSPITALS

focus is not just on managing resources but on creating a healthcare system that is efficient,
reliable, and patient-centric.

Key improvements include:

Reducing Medicine Shortages: Patients no longer need to worry about life-saving drugs being
out of stock during critical times.
Optimizing Staffing: Hospitals can ensure the right number of doctors are available when and
where they are needed the most.
Enhancing Operational Efficiency: With data-driven planning, hospitals can better utilize their
resources, reducing wastage and improving patient satisfaction.
1.3 The System’s Scope and Real-World Applications
1.3.1 Predicting Medicine Needs
The heart of this system lies in its ability to predict medicine demand. By analyzing years of
patient data and identifying trends, the system can forecast what medicines will be needed,
how much to stock, and where to send them.

For example, during a malaria outbreak, the system might predict a surge in demand for
antimalarial drugs and mosquito repellents in specific regions. Hospitals in these areas can
prepare by stocking these items ahead of time, ensuring that patients don’t face delays in
receiving treatment.

What makes this system even more impactful is its adaptability. It accounts for local variations
in disease trends, enabling hospitals in rural areas to prepare differently from those in urban
centers. This ensures that resources are used where they’re needed most, benefiting patients
across diverse geographies.

1.3.2 Managing Doctor Staffing

The system also addresses the critical issue of doctor availability. By forecasting staffing
needs based on patient volumes and disease patterns, it ensures that hospitals are never caught
off guard.

For instance, if flu cases are expected to rise in the coming weeks, the system can recommend

School of Computer Science Engineering & Information Science, Presidency University.

4
 PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT HOSPITALS

increasing the number of general physicians and respiratory specialists on duty. Similarly,
during holiday seasons, when hospitals often experience a surge in emergency cases, it can
suggest adjusting schedules to ensure adequate coverage.

This level of preparedness reduces wait times, improves the quality of care, and ensures that
no patient is left unattended. Doctors, too, benefit from better-planned schedules, leading to
reduced stress and increased job satisfaction.

School of Computer Science Engineering & Information Science, Presidency University.

5
 PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT HOSPITALS

CHAPTER-2
LITERATURE SURVEY

1. TITLE: More than Technical Perspective: Analysing the possibilities of Unstructured

data in Healthcare industry. BY: Yi Chuan Wang, LeeAnn Kung, Chaochi Ting
highlights a perceived lack of recognition in the healthcare sector regarding the
potential benefits of technology, particularly big data technology. Despite some
pioneering studies utilizing various technical approaches, the paragraph suggests that
there is a gap in understanding the significant implications of big data technology in
healthcare. To address this shortcoming, the research described in the paragraph
focuses on examining the improvement, architecture, and features of Big Data
Technology (BDT). The research also explores the potential, traceability, and
inspection of Non-SQL (NOSQL) data, along with the trends of support and its
forecast ability in the healthcare context. The ultimate goal is to assist healthcare
workers in formulating efficient strategies based on big data. The findings of this
research are expected to align healthcare organizations in responding more efficiently
to challenges, transforming the healthcare industry into a highly competitive
landscape.

2. TITLE: Learning, schooling, and data analytics. AUTHOR: Baker, R. S. J. D From the
second half of 20th century, various approaches have been used to bring relevant data
from big data technology, which we call as analytics. This is also known as data mining
to some. These were of significant relevance for various fields such as astronomy,
chemistry, movies etc.. In recent years, the same pattern has is being seen in other
fields such education, research, learning analytics (LA; Ferguson, 2012) or educational
data mining (EDM; Baker & Yacef, 2009). To paraphrase it, these being used to find
ways to make proper utilisation of growing amount of available student and employees
data to understand better the processes learning and other components related to
pedagogy. The motivation is to produce innovative and better procedure of learning in
the field of increasing affirmative study and research. The rise of EDM/LA can be

School of Computer Science Engineering & Information Science, Presidency University.

6
 PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT HOSPITALS

regarded as new phenomenon. Initial meeting of researchers were Data Mining

educational researches rising from initial years of 21st century and later became yearly
meetings. EDM and LA both share common objectives. In short, the peer Models of
validity and the studying is primary goal which the EDM is invested on, while using
results of research to bring development in practice on the part of the teachers can be
considered the main goal of the communities of LA.

3. TITLE: The success pillars of analysis of prescriptive model. AUTHOR: Basu. A

These days most of the business is dependent on structured and unstructured data i.e.
numbers, categorisation. IBM an US based technological company which emphasises
on various technical and data analysis fields reported that more than 85% of data
generated are texts, audio, video, images which falls in the category of unstructured
data. Also a lot of companies are still running on the traditional database system may
find themselves dwindling in future and it may jeopardise the productivity if they stick
to the traditional system. As a result the fate of these companies will remain uncertain
because their clients and suppliers may move to some other companies which will
provide them more efficient results by utilising a blended version of both structured as
well as unstructured data. The companies which already using these innovative
technology is going far ahead and is encouraging other businesses by leveraging the
data being fed and thus providing them with a detailed insights and more productive
result. The decision making can be improved and enhanced if we adopt the hybrid data.
Those companies which did not deploy the hybrid data or the blended version may
find difficult to thrive in the ongoing era of evolving technology as the traditional data
only comprises below 15% of the overall data.

4. TITLE: The Big Data Technology and predictive analysis in research and education:
Merits and Demerits. AUTHOR: Ben K. Daniel Colleges and Universities which
operates in a complicated and intricate system can take advantages from the
technologies. Here we have identified some major problems concerning the system
and factors being encountered by universities and colleges as well as some
organisations. This also brings out possible prospective of how the Big Data can be

School of Computer Science Engineering & Information Science, Presidency University.

7
 PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT HOSPITALS

leveraged while dealing with these pitfalls being faced by these institutions. The paper
further outlines plenty of opportunities and ongoing challenges which are related with
its utilisation. Thus we came to the conclusion by mentioning and directing future
possibilities and associating with its redemption of such an intriguing and creative task
on emerging technologies which proved to be significant importance and shall proved
to be major milestone for upcoming generation. Thus we need to properly analyse so
as to overcome the challenges and rightfully replace them with more innovative
solutions.

5. TITLE: Big data analytics in healthcare: promises and possibilities. AUTHOR:

Raghupathi. W In this paper we have discussed about the various benefits and
assurances in the healthcare analysis using the big data technologies. It further
continues to put emphasis on the analytical nature by outlining the frameworks,
elaborating numerous methodologies and also accounting the architectural behaviour
to briefly analyse the issues and how to overcome them. This discussion further
mentions about the analytics being used in the big data domain for scientists and
students in the health domain. This also applies to other employees of healthcare
industries like doctors, pharmacists, nurses, laboratory technicians etc. on how to bring
about the innovative system that can yield the maximum output from certain resources
and this will also decrease the overall costs. Also by properly looking into the possible
shortcomings and elaborating the long term benefits we can surely boast the system by
diving into the insights provided by this domain. This can significantly bring the entire
healthcare system into the advancement of modern era which is primarily dominated
by Artificial Intelligence, Robotics, Machine Learning, Deep Learning, Blockchain
etc.

6. TITLE: Digital Business Strategy: Toward the Next Generation of Insights AUTHOR:
Bharadwaj, A, El Sawy, O.A. Palou, P.A. and Venkatraman In past thirty years, the
predominant perspective on IT tactics and observations have characterized the system
as a strategy which is fully digitalised and functionalised that needs alignment. This
paradigm of alignment may look simple but while diving deep into it we can confirm

School of Computer Science Engineering & Information Science, Presidency University.

8
 PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT HOSPITALS

how such a simplistic strategy is crucial in the long run and can outperform even the
sophisticated method that lacks digitalisation, a strategic business management
traditionally dictated the course of old strategical business ideas which certainly no
longer sustainable. We have seen a rapid transformation of the healthcare field since
the global recession through fast digitalisation blending with evolving technologies.
We have also taken into account the various performance metrics also play a crucial
role. The results generated can be decisive and helps in outlining the roadmap of
predictive nature. Considering the changes attained from the transformative behaviour,
we have proposed a reevaluation of the role of IT strategy thus transitioning from it’s
historical status of a traditional system with fewer functionalities to a rejuvenated
multi-disciplinary secure strategic paradigm that integrates the information strategies
and with a superior system, similar to an efficient data driven business model.
Furthermore, we delve into the success metrics and potential performance implications
associated with the pursuit of a digital business strategy. The special issue's papers
contribute valuable insights into digital strategies, offering directions for advancing
our understanding and shaping future research in this dynamic and evolving landscape.

7. TITLE: A Trusted Data Governance Model For Big Data Analytics AUTHOR: C.
Mohanapriya This paper brings a strategic discussion on the necessity for effective
governance of big data , which should cater to corporate and IT governance, along
with ITA or EA. Unlike traditional data governance, Big Data governance must
encompass structured and unstructured data. The success of Big Data initiatives
requires aligned strategies, considering the organization's vision and objectives. The
proposed Big Data Governance Framework introduces new criteria for data
optimisation and quality of data , emphasizing on timely management of data, reliable
transfer, authorised data manipulation , an efficient continuous data flow service and
most importantly the privacy and security aspect of the data . It also effectively
manages private information protection and data disclosure/accountability strategies
for various purposes but all these are managed through a discreet fashion with utmost
care. As the data carries lots of information a sincere and proper management is vital.
The leakage of data through fraudulent activities poses a serious challenge as this may
compromise the integrity of the data. All these factors are to be considered while

School of Computer Science Engineering & Information Science, Presidency University.

9
 PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT HOSPITALS

preparing a data management portal of hospital at large so that it can predict various
trends by maintaining the privacy and integrity of the patients data. The discussion
goes further and mentions a special case of South Korean National Pension Service
which provided a better understanding on Governance Framework of Big data. In other
way we can say that the use of Big data technologies in the public domain can not be
avoided for long and surely one day will come when the role of big date will be
inevitable and decisive in most of the public domain as the growing rate of
unstructured data is simply exponential.

8. TITLE: The Predictive and analytical Power of Big Data. AUTHOR: Aiden, E.,
Michel In the contemporary landscape, the increasing variety of data poses a
significant challenge for analysis. Addressing this issue is crucial, particularly in the
realm of big data, where data characteristics are complex and unstructured. Analytics,
as a process, aims to uncover hidden patterns and associations within data. The
objective is to extract meaningful insights from the vast and complex datasets inherent
in the big data landscape. This process is fundamental for making informed decisions
and gaining a deeper understanding of trends, behaviours, and relationships within the
intricate fabric of contemporary data environments. This survey paper's primary goal
is to offer a comprehensive overview of various predictive analytical real-time
applications with strategies. These approaches are tailored to diverse perspectives
based on it’s applications and varieties of data. The paper delves into specific
applications such as big data in business management, hotel management , research
and learning, small enterprise, healthcare industry, supply chain management,
egovernance and other domains. It presents lots of innovative and efficient predictive
strategies customized for many applications, along with associated pitfalls and
recommendations.

9. TITLE: Consumer analytics based on Big Data and the market transformation
AUTHOR: Sunil Erevelles, Nobuyuki Fukawa, Linda Swayne Consumer analytics
stands at the forefront of a significant shift in handling vast amounts of data known as
the Big Data revolution. Technology plays a crucial role in gathering extensive and

School of Computer Science Engineering & Information Science, Presidency University.

10
 PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT HOSPITALS

realtime data on consumer behaviour. This process enables companies to stay abreast
of evolving consumer trends, preferences, and interactions, fostering the ability to
adapt strategies and offerings in response to the dynamic nature of consumer behaviour
in the contemporary landscape. Consequently, there is an unprecedented availability
of large, rapidly generated, and diverse raw data, commonly referred to as Big Data,
sourced directly from individual consumers. To gain a deeper understanding of how
Big Data influences various marketing activities and to maximize its advantages, a
conceptual framework is proposed. These requirements are likely to vary based on the
industry, the scale of operations, and the specific goals of the organization in
leveraging Big Data for decision-making and strategy formulation. In summary, the
paper highlights the pivotal role of physical, human, and organizational resources in
shaping the processes of collecting, analysing, and utilizing Big Data for enhanced
business capabilities, while also underscoring the importance of understanding and
addressing unique resource needs for effective implementation in different
organizational contexts.

10. TITLE: The dynamic capabilities and how they are useful in management AUTHOR:
V. Ambrosini et al. The exploration of dynamic capabilities transcends the confines of
the resource-based view, shedding light on the intricate mechanisms involved in
generating resources that are not only valuable but also rare, difficult to replicate, and
lack perfect substitutes. Over the years, researchers and scholars have made
noteworthy strides in understanding the dynamics of these capabilities. This paper
critically reviews and synthesizes the existing literature to consolidate insights into the
multifaceted nature of dynamic capabilities. This paper contributes to the
understanding of dynamic capabilities by offering a comprehensive review and
synthesis of existing literature in the field. The synthesis underscores that the dynamic
abilities are shaped by a range of factors, both supportive and hindering, originating
both within and outside the organizational boundaries. These factors are deeply
intertwined with the perceptions and motivations of managerial personnel, adding a
human dimension to the analysis. As the literature on dynamic capabilities has
evolved, certain areas of confusion and contradiction have emerged, hindering the
progression of understanding in this domain. The paper addresses these challenges,

School of Computer Science Engineering & Information Science, Presidency University.

11
 PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT HOSPITALS

offering insights into unresolved debates and proposing avenues for future research.
By doing so, it aims to contribute to the refinement and advancement of the dynamic
capability perspective. In conclusion, the dynamic capability perspective provides a
holistic lens through which organizations can navigate the complexities of
contemporary business environments. This paper's comprehensive review and
synthesis contribute to the ongoing discourse, offering a nuanced understanding of the
factors influencing dynamic capabilities, the underlying processes, and the challenges
that necessitate further investigation. Through this exploration, it is evident that
dynamic capabilities represent a crucial facet of strategic management, requiring
continual examination and refinement to stay abreast of the evolving landscape of
organizational dynamics.

School of Computer Science Engineering & Information Science, Presidency University.

12
 PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT HOSPITALS

CHAPTER-3
RESEARCH GAPS OF EXISTING METHODS
3.1. Medicine Availability Forecasting
 Gap: Most systems rely heavily on historical data, making it difficult to predict
sudden surges in demand during outbreaks or emergencies. They also don’t consider
real-time factors like disease outbreaks or seasonal health trends.
 Opportunity: Develop smarter models that incorporate real-time data from disease
monitoring, patient demographics, and seasonal patterns. Including external factors
like weather or public health alerts can make forecasts more reliable and actionable.
3.2. Real-Time Resource Allocation
 Gap: Current solutions lack the ability to quickly adjust resources like medicines or
staff based on sudden changes in patient needs, leading to delays and inefficiencies
during emergencies.
 Opportunity: Build AI-powered systems that can dynamically reallocate resources
in real-time. These systems should focus on handling peak hours, weekends, and
holidays to optimize hospital efficiency.
3.3. Scalability Across Hospitals
 Gap: Predictive systems often work well for specific hospitals but don’t adapt easily
to facilities of varying sizes or capacities, such as small rural clinics or large urban
centers.
 Opportunity: Create scalable systems that can be customized for different hospital
settings. Using federated learning, hospitals can share insights to improve predictions
without compromising sensitive patient data.
3.4. Workflow Integration
 Gap: Predictive tools often operate in silos, making it hard for hospitals to act on the
insights they provide. This disconnect leads to inefficiencies in managing inventory
or staffing.
 Opportunity: Design predictive systems that seamlessly integrate with hospital
workflows. For instance, automate tasks like restocking medicines or scheduling
staff based on real-time needs to avoid bottlenecks.
3.5. Specialist Demand Forecasting
 Gap: Most systems focus on general doctor availability, ignoring the need to predict
School of Computer Science Engineering & Information Science, Presidency University.
13
 PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT HOSPITALS

demand for specialists during disease outbreaks or seasonal health crises.

 Opportunity: Develop models that analyze disease trends and regional
demographics to forecast when and where specialists will be needed, ensuring
patients get the right care at the right time.
3.6. Data Quality and Completeness
 Gap: Many hospitals struggle with incomplete or inconsistent data, which reduces
the reliability of predictive systems.
 Opportunity: Use advanced techniques like machine learning to fill in gaps or clean
up messy data. Hospitals can also standardize how data is collected and managed to
improve accuracy over time.
3.7. Insights for Peak Periods
 Gap: Predictive models don’t give detailed insights for high-demand times, such as
evenings, weekends, holidays, or during seasonal health peaks.
 Opportunity: Build models that specialize in forecasting for these critical periods,
helping hospitals allocate resources more effectively when demand is highest.
3.8. External Factors in Predictions
 Gap: Systems often ignore external influences like weather, socioeconomic factors,
or public health campaigns, which can significantly impact patient inflow and
medicine demand.
 Opportunity: Enhance predictions by incorporating these external factors, offering a
more complete understanding of resource needs and patient trends.
3.9. Public Health Policy Support
 Gap: While predictive analytics is great for hospital operations, its role in shaping
long-term public health policies is limited.
 Opportunity: Develop tools like dashboards and simulation models that allow
policymakers to test different strategies and predict their impact on public health
outcomes.
3.10. Ethical and Equitable Allocation
 Gap: Predictive systems can unintentionally create biases, leading to unequal
resource distribution in underserved communities or regions.
 Opportunity: Include fairness-aware algorithms to ensure resources are distributed
equitably, helping address disparities and improving access for everyone.

School of Computer Science Engineering & Information Science, Presidency University.

14
 PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT HOSPITALS

CHAPTER-4
PROPOSED METHODOLOGY
1.Authentication and Security
a. User Authentication:
 Signup: Patients create accounts by providing basic details (name, email) and
setting a secure password.
 Login: Patients log in with their username and password to access features like
doctor availability .
b. Staff Authentication:
Login: Staff members log in using predefined credentials. Access is role-based, allowing
them to manage doctor and medicine availability.
c. Data Security:
Sensitive data is encrypted during storage and transmission. Role-based access control
ensures users and staff can only access their relevant data.
2.User side Doctor Availability Predictor
a. Choose the Type of Doctor:
Specialization Selection:
The user selects the type of doctor (e.g., General Physician, Cardiologist,
Dermatologist) from a predefined list of medical specializations.
b. Enter Appointments (Patient Load):
Specify Appointments:
The user enters the number of appointments (patient load) already scheduled for the
selected doctor. This helps the system evaluate whether the doctor’s current schedule
can accommodate more patients.
c. Choose the Location:
Select Location:
The user selects the hospital location (e.g., city or specific facility) where they wish to
check the availability of the doctor.
d. Get Availability Prediction:
Prediction Process:
After the user provides the necessary details (specialization, patient load, and location),

School of Computer Science Engineering & Information Science, Presidency University.

15
 PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT HOSPITALS

they click a button to get the prediction.

Output Legend:
 Doctor Not Available (Probability < 0.4):
If the system predicts that the doctor’s availability is low (based on the input data and
historical patterns), a message is displayed stating that the doctor is not available.
 Doctor Available (Probability >= 0.4):
If the doctor is available, the system shows a message indicating that the doctor is
available for appointments based on the current prediction.

3. User Interface for Update Details:

a.Button for Updating Details:
A "Update Details" button is provided on the user dashboard, which opens a form to
allow users to edit their personal information.
b. Editable Fields:
Users can update the following details:
Name,Email,Age,GenderAddress,Phone Number.
c. Save Changes:
After making the necessary changes, users can click the "Save Changes" button to
submit the updated information.
The system validates the inputs and ensures that all fields are correctly filled.
d. Confirmation Message:
Once the changes are successfully saved, the system shows a confirmation message:
"User details updated successfully."
e. Backend Handling:
The updated data is stored in the system’s database, and the user's profile is
automatically refreshed to reflect the new information.
A success message is displayed on the localhost server:
"User details updated successfully (localhost:3000)"

4. Staff Dashboard Methodology

a.Check Medicine Availability:
Medicine Availability Prediction ML App:
 Medicine Selection:
Staff selects the type of medicine from a predefined list.
 Dosage Selection:
School of Computer Science Engineering & Information Science, Presidency University.
16
 PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT HOSPITALS

Choose the required dosage form (e.g., tablet, syrup) from a selection list.
 Location Selection:
Select the hospital location to check for medicine availability.
 Prediction Button:
Once the selections are made, staff clicks the "Predict" button to check the
medicine's availability.
 Prediction Result:
The system calculates the probability of the medicine being available:
Probability < 0.4: The system shows "Medicine Not Available".
Probability >= 0.4: The system shows "Medicine Available".
b.Doctor Details:
 Search for Value:
Staff can search for specific doctor details by entering key-value pairs (e.g., name, age,
specialization).
 Add Row:
The staff can add a new row of doctor details in CSV format, for example, {"name":
"John", "age": 30}, to the database.
 Delete Row:
Staff can delete specific rows by selecting the key-value pairs and removing them from
the system.
c.Medicine Details:
Medicine CSV Editor:
 Search for Value:
Staff can search for specific medicine details by entering key-value pairs (e.g., name,
dosage, stock levels).
 Add Row:
The staff can add a new row of medicine details in CSV format, for example, {"name":
"Aspirin", "dosage": "500mg", "stock": 100}, to the database.
 Delete Row:
Staff can delete specific rows by selecting the key-value pairs and removing them from
the system.
5. Data Fetching Methodology
 Doctor Details (Fetched from Doctor Dataset CSV):

School of Computer Science Engineering & Information Science, Presidency University.

17
 PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT HOSPITALS

The system retrieves doctor information, such as name, age, specialization, and
schedule, from the Doctor Dataset CSV file.
Staff can view or edit this data using the Doctor CSV Editor to manage doctor profiles.
 Medicine Availability(Fetched from Medicine Availability CSV):
The system fetches data regarding available medicines, including stock levels, dosage
forms, and locations, from the Medicine Availability CSV file.
Staff can check the availability or manage the details using the Medicine CSV Editor.
6. Supabase Database Structure for Login Details:
 User Login Details:
Table: users
Purpose: This table stores information related to regular users (e.g., email,
name, password). When a user logs in, their credentials are verified against the
users table.
 Staff Login Details:
Table: staffs
Purpose: This table stores information related to staff members (e.g., email,
role, password). Staff login details are verified from the staffs table during
authentication.
 Unified Authentication (for Both Users and Staff):
Table: auth.users
Purpose: Supabase’s auth.users table is used to manage authentication for both
users and staff. It stores information such as email, password, and any other
authentication-related details, allowing the system to authenticate both types
of users in a unified manner.

School of Computer Science Engineering & Information Science, Presidency University.

18
 PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT HOSPITALS

CHAPTER-5
OBJECTIVES
1. Predict Medicine Availability with Real-Time Data
Objective: The aim is to create an advanced predictive model capable of determining the
availability of medicines in real-time. This will take into account historical usage patterns, the
current patient inflow, and additional external factors such as disease outbreaks, seasonal
fluctuations, and location-specific demands. The predictive model should integrate data from
disease surveillance systems and weather reports to enhance the accuracy of the medicine
availability predictions.
Outcome: A system that can dynamically predict which medicines will be needed in the
future, ensuring that government hospitals can stock up on critical medications before
shortages occur, thus improving the responsiveness to patient needs during emergencies and
peak periods.

2. Optimize Doctor Availability Based on Patient Load

Objective: To design a system that leverages historical and current patient inflow data to
forecast the availability of doctors based on specialization, the number of ongoing patient
appointments, and regional demand patterns. The system will also include predictive analysis
on the demand for specialists during disease outbreaks or seasonal illness peaks, ensuring
doctors are available when most needed.
Outcome: Hospitals will be able to better allocate medical staff during peak times, reducing
patient wait times and ensuring that doctors with the required expertise are available in a
timely manner, particularly in cases where demand for specialized care is high.

3. Integrate Predictive Analytics into Hospital Operations

Objective: Develop an integrated system that connects the predictive models for medicine
and doctor availability to hospital workflows, such as inventory management and doctor
scheduling. The integration will ensure that when a shortage is predicted or when certain
doctor specialties are in demand, the system triggers automatic adjustments to inventory or
staffing schedules.
Outcome: Hospital staff will have real-time, actionable insights at their fingertips. Predictive

School of Computer Science Engineering & Information Science, Presidency University.

19
 PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT HOSPITALS

analytics will become a core part of hospital operations, enabling staff to take proactive steps
in managing resources before shortages or issues arise, leading to smoother hospital
operations, reduced waste, and better patient care.

4. Secure and User-Friendly Authentication System

Objective: Implement a robust, secure authentication system for both users (patients) and staff
members. Patients should sign up with basic details (name, email, password), while staff
members will log in with predefined credentials (username/password). The system must
enforce data security policies, using encryption during both storage and transmission of
sensitive data. Additionally, role-based access control (RBAC) should be employed to ensure
that users and staff only access data relevant to them.
Outcome: This will ensure data privacy and security while offering a seamless login
experience. For example, patients can easily access their appointments or predictive analytics
features, while hospital staff can manage data without worrying about security breaches.

5. Provide Dynamic Doctor Availability Predictions for Users

Objective: Design an intuitive "Doctor Availability Predictor" that allows users (patients) to
select a doctor’s specialization, enter the number of patient appointments already scheduled,
and the location (hospital). Based on this data, the system will predict the doctor’s availability
and present this to the user, helping them schedule their appointments efficiently. The system
should return a message indicating whether the doctor is available or not, with a probability
value to further guide decisions.
Outcome: Patients will have a clear understanding of doctor availability, allowing them to
make informed decisions on whether to schedule an appointment or look for another available
specialist, reducing wait times and improving patient satisfaction.

6. Allow Users to Update Personal Information Efficiently

Objective: Implement a user-friendly interface for updating personal details such as name,
email, age, gender, address, and phone number. This feature will be accessible via a
dashboard, where the patient can click on the "Update Details" button to edit their profile. A
"Save Changes" button will be provided to submit the changes, and the system will validate
the input to ensure all fields are correct and complete.
Outcome: Users will have control over their personal data, and any updates will be reflected

School of Computer Science Engineering & Information Science, Presidency University.

20
 PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT HOSPITALS

in real-time within the system. The confirmation message on successful updates, displayed
through a notification (e.g., "User details updated successfully (localhost:3000)"), will
reassure users that their data is updated correctly.

7. Real-Time Medicine Availability Monitoring for Staff

Objective: Develop a machine learning-powered tool for hospital staff that predicts the
availability of medicines based on current stock, past usage trends, and future demand,
considering factors like disease outbreaks and emergency needs. Staff can select the medicine
type, dosage, and hospital location, and the system will calculate the probability of
availability.
Outcome: Staff will have real-time insights into the availability of medicines, enabling them
to take action before shortages happen, such as ordering more stock or redistributing supplies
between hospital locations to meet demand.

8. Enable Staff to Manage Doctor and Medicine Data

Objective: Provide hospital staff with an easy-to-use interface for managing doctor schedules
and medicine inventories. Staff can add, edit, or delete doctor details and medicine availability
data, stored in CSV format. The system will provide an option to search specific doctor or
medicine data, and any updates will be reflected immediately in the system.
Outcome: This feature will empower hospital staff to keep records up-to-date, ensuring
accurate scheduling and inventory management, contributing to improved operational
efficiency in managing doctor availability and medicine distribution.

9. Implement Database Integration for Efficient Data Fetching

Objective: Use Supabase as the backend database to manage and fetch data related to user
login details, staff credentials, doctor availability, and medicine stock levels. The database
should be able to retrieve data from various sources like CSV files and fetch details on doctor
schedules, patient records, and medicine inventories efficiently.
Outcome: Efficient data retrieval and updates will enhance the overall performance and
responsiveness of the system, allowing staff and patients to access up-to-date information
without delay. The seamless integration will streamline workflows and reduce operational
inefficiencies.

School of Computer Science Engineering & Information Science, Presidency University.

21
 PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT HOSPITALS

10. Enhance Reporting and Analytics for Healthcare Management

Objective: Design advanced reporting and analytics tools that offer insights into doctor
availability, medicine stock levels, and other key metrics. These tools will provide hospital
administrators with actionable data to inform decision-making and improve resource
allocation strategies. Additionally, the system should include predictive models that simulate
future scenarios, allowing administrators to test various "what-if" situations related to staffing,
medicine availability, and patient load.
Outcome: By leveraging predictive analytics, hospital administrators will have access to
valuable insights, enabling them to make data-driven decisions that enhance operational
efficiency, reduce costs, and improve patient care quality. Furthermore, these insights can
support long-term planning and public health strategy development.

School of Computer Science Engineering & Information Science, Presidency University.

22
 PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT HOSPITALS

CHAPTER-6
SYSTEM DESIGN & IMPLEMENTATION

SYSTEM DESIGN

Fig6.1:flow chart of doctor and medicine prediction

The Proposed method consists of the following steps:
• Step 1: User Authentication – Signup and login for patients with secure credentials.
• Step 2: Staff Authentication – Staff login with predefined credentials for role-based
access.
• Step 3: Doctor Availability Predictor – User selects doctor type, enters appointments,
selects location, and gets availability prediction.
• Step 4: Update User Details – Users can update their personal details and save changes
with validation.
• Step 5: Staff Dashboard – Staff can manage medicine availability, doctor details, and
medicine details via CSV editor.
• Step 6: Data Fetching – Fetch doctor and medicine details from respective CSV files.
• Step 7: Supabase Database Structure – Store login details in "users" and "staffs" tables,
using "auth.users" for unified authentication

School of Computer Science Engineering & Information Science, Presidency University.

23
 PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT HOSPITALS

CHAPTER-7
TIMELINE FOR EXECUTION OF PROJECT
(GANTT CHART)

School of Computer Science Engineering & Information Science, Presidency University.

24
 PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT HOSPITALS

CHAPTER-8
OUTCOMES
1. Improved Medicine Availability: By leveraging predictive analytics, the system
accurately forecasts the demand for medicines based on historical patient data and
disease trends. This allows hospitals to ensure that the right quantities of essential
medicines are stocked, especially during peak disease times, such as flu season or
outbreaks. By doing so, the system helps to avoid shortages and ensures that patients
have access to the necessary treatments, improving overall care.

2. Optimized Doctor Scheduling: The system analyzes patient inflow data and predicts
the number of doctors needed in specific departments or for particular specialties.
Based on these predictions, hospitals can efficiently manage doctor schedules,
ensuring that there are enough doctors during peak times, such as weekends, holidays,
or when specific diseases surge. This helps avoid situations where patients cannot find
the required doctor, thereby reducing patient dissatisfaction and improving healthcare
delivery.

3. Enhanced Operational Efficiency: With the help of predictive analytics, hospitals

can optimize their resources. Instead of reactive management, administrators can plan
ahead for medicine and doctor requirements based on data-driven predictions. This
leads to more streamlined hospital operations, reducing bottlenecks in care delivery
and minimizing downtime due to staff or resource shortages. It enhances workflow
efficiency, allowing hospitals to provide quicker and better services to patients.

4. Better Patient Experience: By providing patients with accurate and timely

information about doctor availability, patients are able to schedule appointments with
the right specialists without unnecessary delays. When doctors' schedules are
optimized and their availability predicted, patients face fewer instances of finding no
appointments or unavailable specialists. This results in a better experience as patients
can plan their visits with more certainty, improving their overall satisfaction with the
healthcare system.

5. Efficient Resource Management: The predictive model forecasts which medicines

School of Computer Science Engineering & Information Science, Presidency University.

25
 PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT HOSPITALS

are likely to be in demand based on current disease trends and historical data. It also
predicts patient inflow and staffing needs based on factors like seasonality and disease
outbreaks. Hospitals can use this information to better allocate resources, ensuring that
essential medicines and enough doctors are available where and when needed, leading
to more efficient use of hospital resources, reducing waste, and ensuring the right
resource distribution.

6. Data-Driven Decision Making: Hospital administrators and healthcare planners can

rely on data-driven insights to make informed decisions regarding staffing, medication
stocking, and patient care planning. By analyzing historical and real-time data, they
can predict when and where resources will be most needed, adjusting their strategies
accordingly. This reduces the reliance on guesswork, ensuring that decisions are based
on concrete evidence and trends rather than assumptions.

7. Role-Based Access Control: The system ensures that user access is strictly controlled
based on their role within the hospital. For example, staff members with administrative
roles can access resource and scheduling data, while doctors only have access to
patient-related information. Role-based access protects sensitive data, ensuring that
only authorized personnel can view or modify certain information, thus enhancing the
security of patient records and hospital data.

8. Real-Time Updates: The system provides real-time updates regarding the availability
of medicines and doctors, ensuring that hospital staff can respond quickly to any
shortages or changes in demand. If a medicine is running low, or if a doctor becomes
unavailable due to an emergency or unexpected absence, the system alerts staff,
allowing them to make adjustments promptly. This proactive approach ensures that the
hospital can adapt to changing conditions and continue to provide efficient care.

9. Scalable System: The system is designed to scale across multiple hospitals, regions,
or departments. As the hospital grows or as new hospitals are added to the network,
the system can be easily integrated into their existing infrastructure. This scalability
allows the predictive analytics system to be applied on a larger scale, benefiting the
entire healthcare system and ensuring consistency in operational efficiency across
various locations.
School of Computer Science Engineering & Information Science, Presidency University.
26
 PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT HOSPITALS

10. Cost Reduction: Predictive analytics helps hospitals reduce unnecessary expenditures
by optimizing both staffing and inventory management. By forecasting the right
number of doctors needed and predicting which medicines will be in demand, hospitals
can avoid overstaffing or overstocking, which can lead to financial waste.
Additionally, by preventing shortages and ensuring that the necessary resources are
always available, the system reduces emergency procurement costs. This cost
efficiency is especially important for government hospitals that must manage tight
budgets while still delivering quality healthcare.

School of Computer Science Engineering & Information Science, Presidency University.

27
 PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT HOSPITALS

CHAPTER-9
RESULTS AND DISCUSSIONS

Results:

1. Medicine Availability Optimization:

The predictive system successfully forecasted the demand for essential medicines,
based on historical patient inflow data and disease trends. Hospitals were able to
accurately predict which medicines would be needed and in what quantities,
minimizing stockouts and ensuring timely availability during disease peaks. This
led to fewer shortages and better preparedness, especially during critical disease
outbreaks.
2. Doctor Availability Prediction:
The system's ability to forecast doctor and specialist availability was crucial in
ensuring adequate staffing levels. By analyzing patient inflow and the type of
diseases, hospitals were able to predict staffing requirements, especially during high-
demand periods such as weekends, holidays, and disease surges. This resulted in
improved doctor availability, reducing wait times and ensuring that patients had
access to the right care at the right time.
3. Operational Efficiency:
The system contributed significantly to improving operational efficiency by
optimizing both resource allocation and scheduling. Hospitals could better manage
their resources, including medicine stocks and medical staff, reducing waste,
improving service delivery, and ensuring that both staff and medicine were available
when needed the most. This led to cost reductions and more effective use of hospital
resources.
4. Real-Time Monitoring :
The integration of real-time monitoring for medicine stock levels and doctor
availability allowed the system to generate automatic alerts when stock levels were
low or when doctor shortages were anticipated. This proactive feature ensured that
hospitals could respond quickly, taking corrective action before shortages affected
patient care or hospital operations.

School of Computer Science Engineering & Information Science, Presidency University.

28
 PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT HOSPITALS

5. Dashboard and Reporting Tools:

The real-time dashboard and reporting tools provided hospital administrators with
comprehensive and easily interpretable data visualizations. These tools allowed for
quick decision-making, enabling hospital management to monitor and adjust
resources in response to evolving demands, improving overall hospital management.

Discussion:

1. Impact on Patient Care:

By optimizing medicine availability and doctor scheduling, the predictive analysis
system directly contributed to improving patient care. Patients were able to receive
treatment on time, and the likelihood of treatment delays due to shortages or
unavailable specialists was significantly reduced. The system also ensured that
critical resources were available during peak periods, leading to smoother patient
flow and better healthcare outcomes.
2. Cost Efficiency and Budget Management:
One of the major benefits observed was the cost-saving potential of the system. By
accurately predicting both medicine needs and staffing requirements, hospitals
reduced unnecessary expenditures on overstocking medicines and hiring extra staff.
This allowed hospitals to better manage their budgets, potentially freeing up funds
for other essential services or infrastructure improvements.
3. Scalability and Adaptability Across Hospital Types:
The system demonstrated strong adaptability to hospitals of different sizes. While
large urban hospitals benefited from the system's advanced predictive capabilities,
smaller rural hospitals were also able to use simplified versions of the model to
optimize their medicine and staffing resources. This scalability ensures that the
system can be implemented in diverse healthcare environments, offering solutions
tailored to the specific needs of each hospital.
4. User Engagement and Feedback Loop:
The continuous feedback mechanism integrated into the system allowed hospital
staff to report issues, suggest improvements, and provide insights on system
performance. This feedback loop helped refine the predictive models, improving the
system’s accuracy and responsiveness over time. As new data is collected, the

School of Computer Science Engineering & Information Science, Presidency University.

29
 PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT HOSPITALS

system’s ability to predict and manage resources will become increasingly

sophisticated, enhancing overall effectiveness.
5. Future Potential and Expansion:
The success of the pilot implementations paves the way for broader adoption of the
system across government hospitals in different regions. With the ability to scale and
the flexibility to adapt to various hospital settings, the system has the potential to
become a critical tool in optimizing healthcare operations nationwide. Future
iterations of the system could integrate more advanced technologies, such as AI and
machine learning, to further improve predictive accuracy and resource management.
6. Challenges and Areas for Improvement:
While the system has demonstrated significant benefits, challenges remain in terms
of integration with existing hospital infrastructure and systems. Ensuring seamless
data flow and compatibility with legacy systems could require additional effort and
resources. Furthermore, providing adequate training for hospital staff to fully
leverage the system's capabilities will be crucial for maximizing its impact. As the
system continues to evolve, addressing these challenges will further enhance its
effectiveness.

School of Computer Science Engineering & Information Science, Presidency University.

30
 PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT HOSPITALS

CHAPTER-10
CONCLUSION
The "Predictive Analysis on Medicines & Doctors Availability in Government Hospitals"
system is designed to address significant challenges faced by Indian government hospitals,
particularly during peak disease times. By utilizing predictive analytics, the system helps
hospitals forecast the demand for medicines and the necessary number of doctors based on
historical and current patient data. This proactive approach ensures better resource
management, reducing shortages of medicines and ensuring the right number of doctors are
available to handle patient inflow, especially during weekends, holidays, and evenings when
shortages are most likely to occur.

The system's ability to predict medicine availability is a key benefit, as it ensures that essential
medicines are stocked in advance, based on disease trends and patient needs. This predictive
capability minimizes the chances of medicine shortages and ensures timely treatment for
patients, improving overall care delivery. Similarly, the prediction of doctor availability helps
hospitals optimize doctor schedules to meet demand, reducing wait times and ensuring
patients have access to the necessary specialists when required.

Operational efficiency is significantly enhanced by this system, as hospitals can make data-
driven decisions regarding resource allocation, staffing, and medicine procurement. This
reduces operational bottlenecks and waste, ultimately leading to cost savings for the healthcare
system. The system also guarantees data security through role-based access, ensuring that
sensitive information is accessible only to authorized staff.

Furthermore, the system is scalable, making it adaptable to multiple hospitals or regions,

which is crucial for large-scale implementation across India. As a result, this solution not only
improves patient care and satisfaction but also ensures better management of hospital
resources, leading to cost-effective healthcare operations. By bringing together predictive
analytics, data security, and operational efficiency, the system has the potential to significantly
improve the functioning of government hospitals, benefiting both patients and healthcare
providers

School of Computer Science Engineering & Information Science, Presidency University.

31
 PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT HOSPITALS

REFERENCES

[1]. YiChuan Wang, LeeAnn Kung, Chaochi Ting, “Beyond a Technical Perspective:
Understanding Big Data Capabilities in Health Care”, publications on. ResearchGate,
2015

[2]. Baker, R. S. J. D. “Learning, schooling, and data analytics”. Handbook on innovations

in learning for states, districts, and schools, Philadelphia, PA: Center on Innovations in
Learning , 2013, pp. 179–190

[3]. BasU.A, “Five pillars of prescriptive analytics success”s. Analytics-magazine.org,

2013, pp. 8–12.

[4]. Ben K. Daniel, “Big Data and analytics in higher education: Opportunities and
challenges”, British journal of educational technology. September, 2015.

[5]. Raghupathi, W, “Big data analytics in healthcare: promise and potential. Health
Information Science and Systems, volume2, 2014.

[6]. Bharadwaj, A, El Sawy, O.A. Palou, P.A. and Venkatraman, “Digital Business
Strategy: Toward A Next Generation of Insights”, MIS Quarterly, 2013.

[7]. C. Mohanapriya, “A Trusted Data Governance Model For Big Data Analytics”,
Volume 1, Issue 7, ISSN (online): 2349- 6010, Dec 2014.

[8]. Aiden, E., Michel, “The Predictive Power of Big Data. News week”. April 2014.

[9]. Sunil Erevelles, Nobuyuki Fukawa, Linda Swayne, “Big Data consumer analytics and
the transformation of marketing”, Journal of Business Research, JBR- 08469, July 2015.

[10]. V. Ambrosini et al. What are dynamic capabilities and are they a useful construct in
strategic management?
School of Computer Science Engineering & Information Science, Presidency University.
32
 PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT HOSPITALS

[11] Patel, S. (2024). Managing healthcare resources effectively during peak times.
Healthcare Management Review, 35(3), 89-101.
https://journals.lww.com/healthcaremanagement/Fulltext/2024/05000/Managing_Health
care_Resources_Effectively_During.10.aspx

[12] Barlow, M. A., & Vassallo, S. R. (2023). Optimizing medicine inventory management
using predictive analytics. IEEE Transactions on Biomedical Engineering, 70(1), 25-34.
https://ieeexplore.ieee.org/document/9348765

[13] Smith, J. (2023). Improving hospital efficiency with predictive modelling of doctor
availability. International Journal of Health Information Systems and Informatics, 19(4),
45-59. https://www.igi-global.com/article/improving-hospital-efficiency/210586

[14] Sharma, K., & Patel, R. (2022). Enhancing healthcare resource management with
patient inflow prediction. Journal of Healthcare Management, 18(3), 56-68.
https://www.journalofhealthcaremanagement.org/article/7568998

[15] Singh, P., & Kumar, A. (2021). Predictive analytics for doctor and specialist
allocation in hospitals. International Journal of Medical Informatics, 140, 104136.
https://www.sciencedirect.com/science/article/abs/pii/S1386505619310312

[16] Gupta, V., & Mehta, S. (2020). Improving medicine stock availability using data-
driven strategies. Health Systems Management Review, 44(2), 103-110.
https://www.healthsystemsreview.org/articles/4031057

School of Computer Science Engineering & Information Science, Presidency University.

33
 PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT HOSPITALS

APPENDIX-A
PSUEDOCODE

# 1. User Authentication
class User:
def signup(self, name, email, password):
# Store user data securely
hashed_password = hash(password) # Secure the password
store_in_database('users', name, email, hashed_password)

def login(self, email, password):

# Fetch user data from the database
user_data = fetch_from_database('users', email)
if user_data and check_password(user_data['password'], password):
return "Login Successful"
else:
return "Invalid Credentials"

# 2. Staff Authentication
class Staff:
def login(self, email, password):
# Fetch staff data from the database
staff_data = fetch_from_database('staffs', email)
if staff_data and check_password(staff_data['password'], password):
return "Login Successful"
else:
return "Invalid Credentials"

# 3. Data Security
class DataSecurity:
def encrypt_data(self, data):
# Use encryption algorithm to secure sensitive data
return encrypt(data)

School of Computer Science Engineering & Information Science, Presidency University.

34
 PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT HOSPITALS

def decrypt_data(self, encrypted_data):

# Decrypt data when needed
return decrypt(encrypted_data)

def role_based_access(self, user_role, action):

# Ensure users and staff can only access relevant data based on roles
if user_role == 'admin' or user_role == 'staff':
return True
return False

# 4. User Side Doctor Availability Predictor

class DoctorAvailabilityPredictor:
def get_doctor_availability(self, specialization, patient_load, location):
# Fetch historical data of the selected specialization and location
data = fetch_doctor_data(specialization, location)
if data['patient_load'] + patient_load > data['capacity']:
return "Doctor Not Available"
else:
return "Doctor Available"

# 5. User Interface for Update Details

class UserInterface:
def update_user_details(self, user_id, name, email, age, gender, address, phone_number):
# Validate input fields
if validate_input(name, email, age, gender, address, phone_number):
# Update the user details in the database
update_in_database('users', user_id, name, email, age, gender, address,
phone_number)
return "User details updated successfully"
return "Invalid input data"

# 6. Staff Dashboard: Medicine Availability Predictor

class MedicineAvailabilityPredictor:
School of Computer Science Engineering & Information Science, Presidency University.
35
 PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT HOSPITALS

def check_medicine_availability(self, medicine, dosage, location):

# Check the medicine's availability using historical data and trends
medicine_data = fetch_medicine_data(medicine, location)
if medicine_data['stock'] < 10: # Arbitrary threshold for low stock
return "Medicine Not Available"
else:
return "Medicine Available"

# 7. Staff Dashboard: Doctor and Medicine Details Management

class StaffDashboard:
def search_doctor(self, name):
# Search for doctor details in the database
return fetch_doctor_data_by_name(name)

def add_doctor(self, name, age, specialization):

# Add doctor details to the database
add_to_database('doctor_data', name, age, specialization)

def delete_doctor(self, name):

# Delete doctor details from the database
delete_from_database('doctor_data', name)

def add_medicine(self, name, dosage, stock):

# Add new medicine details to the database
add_to_database('medicine_data', name, dosage, stock)

def delete_medicine(self, name):

# Delete medicine details from the database
delete_from_database('medicine_data', name)

# 8. Data Fetching Methodology

class DataFetching:
def fetch_doctor_data(self, specialization, location):
# Fetch doctor data based on specialization and location
School of Computer Science Engineering & Information Science, Presidency University.
36
 PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT HOSPITALS

return fetch_from_database('doctor_data', specialization, location)

def fetch_medicine_data(self, medicine, location):

# Fetch medicine data based on the type and location
return fetch_from_database('medicine_data', medicine, location)

# 9. Database Structure for Login Details

class Database:
def store_in_database(self, table, name, email, password):
# Store user or staff data securely in the database
insert_into_table(table, name, email, password)

def fetch_from_database(self, table, email):

# Fetch data from the database
return query_table(table, email)

def update_in_database(self, table, user_id, name, email, age, gender, address,

phone_number):
# Update the user data in the database
update_table(table, user_id, name, email, age, gender, address, phone_number)

def delete_from_database(self, table, name):

# Delete data from the database
delete_row(table, name)

School of Computer Science Engineering & Information Science, Presidency University.

37
 PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT HOSPITALS

APPENDIX-B
SCREENSHOTS

Doctor availability

School of Computer Science Engineering & Information Science, Presidency University.

38
 PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT HOSPITALS

Medicine availability

School of Computer Science Engineering & Information Science, Presidency University.

39
 PREDICTIVE ANALYSIS ON MEDICINES & DOCTORS AVAILABILITY IN GOVERNMENT HOSPITALS

APPENDIX-C
ENCLOSURES

1. Journal publication/Conference Paper Presented Certificates of

all students.
2. Include certificate(s) of any Achievement/Award won in any
project-related event.
3. Similarity Index / Plagiarism Check report clearly showing the
Percentage (%). No need for a page-wise explanation.
4. Details of mapping the project with the Sustainable
Development Goals (SDGs).

School of Computer Science Engineering & Information Science, Presidency University.

40