SIDDAGANGA INSTITUTE OF TECHNOLOGY, TUMAKURU-572103

(An Autonomous Institute under Visvesvaraya Technological University, Belagavi)

A Miniproject Report on

“Real-Time Route Optimisation and Safety System

using Weather Data and Explainable AI”

submitted in partial fulfillment of the requirement for the completion of

V semester of
BACHELOR OF ENGINEERING
in
INFORMATION SCIENCE & ENGINEERING
Submitted by

Puneeth C S 1SI22IS071
Sathvik U S 1SI22IS083
Siddharoodha 1SI22IS095
Suhas K N 1SI22IS105

under the guidance of

Dr.Hemavathi
Asst. Professor
Department of ISE
SIT, Tumakuru-03

DEPARTMENT OF INFORMATION SCIENCE & ENGINEERING

2024-25
 SIDDAGANGA INSTITUTE OF TECHNOLOGY, TUMAKURU-572103
(An Autonomous Institute under Visvesvaraya Technological University, Belagavi)

CERTIFICATE

Certified that the miniproject work entitled “Real-Time Route Optimisation and
Safety System using Weather Data and Explainable AI” is a bonafide work
carried out by Puneeth C S (1SI22IS071), Sathvik U S (1SI22IS083) Siddharoodha M
J (1SI22IS095) Suhas K N (1SI22IS105) in partial fulfillment for the completion of V
Semester of Bachelor of Engineering in Information Science and Engineering from Sidda-
ganga Institute of Technology, an autonomous institute under Visvesvaraya Technological
University, Belagavi during the academic year 2024-25. It is certified that all correc-
tions/suggestions indicated for internal assessment have been incorporated in the report
deposited in the department library. The project report has been approved as it satisfies
the academic requirements in respect of project work prescribed for the completion of V
semester of Bachelor of Engineering degree.

Dr.Hemavathi Head of the Department

Asst.Professor Dr. R Aparna
Project guide Professor
Dept. of ISE Dept. of ISE
SIT,Tumakuru-03

External viva:
Names of the Examiners Signature with date
1.
2.
Real-Time Route Optimization with Weather XAI 2024-25

Dept.of ISE, S.I.T.,Tumakuru-03 2

 ACKNOWLEDGEMENT

We offer our humble pranams at the lotus feet of His Holiness, Dr. Sree Sree Sivaku-
mara Swamigalu, Founder President and His Holiness, Sree Sree Siddalinga Swami-
galu, President, Sree Siddaganga Education Society, Sree Siddaganga Math for bestowing
upon their blessings.

We deem it as a privilege to thank Late Dr. M N Channabasappa, Director, SIT,

Tumakuru, Dr. Shivakumaraiah, CEO, SIT, Tumakuru, and Dr. S V Dinesh, Prin-
cipal, SIT, Tumakuru for fostering an excellent academic environment in this institution,
which made this endeavor fruitful.

We would like to express our sincere gratitude to Dr. R Aparna, Professor and Head,
Department of ISE, SIT, Tumakuru for her encouragement and valuable suggestions.

We thank our guide Dr.Hemavathi, Asst.Professor, Department of Information Science

& Engineering, SIT, Tumakuru for the valuable guidance, advice and encouragement.

Puneeth C S 1SI22IS071
Sathvik U S 1SI22IS083
Siddharoodha 1SI22IS095
Suhas K N 1SI22IS105
 Course Outcomes

After successful completion of mini project, graduates will be able to

CO1: To identify a problem through literature survey and knowledge of contemporary
engineering technology.
CO2: To consolidate the literature search to identify issues/gaps and formulate the engi-
neering problem
CO3: To prepare project schedule for the identified design methodology and engage in
budget analysis, and share responsibility for every member in the team
CO4: To provide sustainable engineering solution considering health, safety, legal, cul-
tural issues and also demonstrate concern for environment
CO5: To identify and apply the mathematical concepts, science concepts, engineering and
management concepts necessary to implement the identified engineering problem
CO6: To select the engineering tools/components required to implement the proposed
solution for the identified engineering problem
CO7: To analyze, design, and implement optimal design solution, interpret results of ex-
periments and draw valid conclusion
CO8: To demonstrate effective written communication through the project report, the
one-page poster presentation, and preparation of the video about the project and the four
page IEEE/Springer/ paper format of the work
CO9: To engage in effective oral communication through power point presentation and
demonstration of the project work.
CO10:To demonstrate compliance to the prescribed standards/ safety norms and abide
by the norms of professional ethics.
CO11: To perform in the team, contribute to the team and mentor/lead the team
Real-Time Route Optimization with Weather XAI 2024-25

CO-PO Mapping

PO1 PO2 PO3 PO4 PO5 PO6 PO7 PO8 PO9 PO10 PO11 PSO1 PSO2
CO-1 3 3 3 3 3 3
CO-2 3 3 3 3 3
CO-3 3 3 3 3
CO-4 3 3 3 3 3 3
CO-5 3 3 3
CO-6 3 3 3 3
CO-7 3 3 3 3
CO-8 3 3 3
CO-9 3 3 3 3
CO-10 3 3 3 3
CO-11 3 3 3 3
Average 3 3 3 3 3 3 3 3 3 3 3 3

PSO mapping to be done by respective Dept.

Attainment level: - 1: Slight (low) 2: Moderate (medium) 3: Substantial (high)
POs: PO1: Engineering knowledge, PO2: Problem analysis, PO3:Design of solutions,
PO4:Conduct investigations of complex problems, PO5: Engineering tool usage, PO6:Engineer
and the world, PO7:Ethics, PO8:Individual and collaborative work, PO9:communication,
PO10:project management and finance,PO11: Life-long learning.

PS01:Computing System, PS02: Communication and Security, PS03: Information Man-

agement

Dept.of ISE, S.I.T.,Tumakuru-03 2

Abstract
This project aims to develop an advanced navigation tool designed to optimize route se-
lection by incorporating real-time data on road and weather conditions. By analyzing
key factors such as road quality, and weather patterns, the tool will provide users with
recommendations for the safest, most efficient, and reliable routes. The primary objective
is to enhance travel safety, reduce travel time, and improve overall route dependability.
This solution has the potential to benefit individual travelers, logistics companies, and
emergency services by ensuring smarter route planning and better decision-making
This project focuses on route optimization by developing a tool that selects the best routes
using real-time data on road quality, and weather conditions. It aims to enhance safety
and reduce travel time by avoiding hazards and congestion through smart routing. By
integrating road and weather data, the tool ensures accurate and reliable route sugges-
tions, even in dynamically changing conditions.

i
Contents
Abstract i

List of Figures ii

List of Tables iii

1 Introduction 1
1.1 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Objective of the project . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.3 Organisation of the report . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

2 Literature Survey 5
2.1 Literature Review on Explainable AI . . . . . . . . . . . . . . . . . . . . . 5
2.2 Literature Review on Route Optimisation . . . . . . . . . . . . . . . . . . 7
2.3 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

3 System Requirements 10
3.1 Software: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
3.2 Hardware . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
3.3 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

4 Proposed Methodology 12
4.1 Introduction about Methodology . . . . . . . . . . . . . . . . . . . . . . . 12
4.2 System Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
4.3 System Workflow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
4.4 Implementation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
4.5 Performance Measurement . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

5 Results 18
5.1 Snapshots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
5.2 Comprehensive Analysis of Route XAI Project . . . . . . . . . . . . . . . . 21
5.2.1 Code Metrics Analysis . . . . . . . . . . . . . . . . . . . . . . . . . 21
6 Conclusion 23

7 Future Work 25

Bibliography 26
List of Figures
4.1 System Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
4.2 System Workflow for the Real Time Route Optimization with Weather &
XAI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

5.1 User Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

5.2 Road Condition Reports . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
5.3 Route Explorer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
5.4 Road Condition Feedback Form . . . . . . . . . . . . . . . . . . . . . . . . 20
5.5 Pothole Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
5.6 Live Navigation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

ii
List of Tables
5.1 File Size Distribution and Functionality Analysis . . . . . . . . . . . . . . 22

iii
Chapter 1
Introduction
Safe and efficient transport by road is a primary prerequisite in most contemporary soci-
eties, but a number of times one does experience the disturbance of services due to unfa-
vorable weather conditions, road closures, or poor formulations in infrastructure projects.
In turn, real-time changes elude classical navigation and often result to longer journey
time, increased risks of accidents, and exacerbation of users’ frustrations. The project
titled “Real-Time Route Optimization and Safety System Using Weather Data and Ex-
plainable AI (XAI)” addresses these issues by introducing smarter, facilitative solutions
that aim at innovation of navigation systems. The system works in real time with weather
data, machine learning algorithms that continuously track road and environmental con-
ditions in order to recommend productive and favorable routes to the user. This new
process of execution with Explainable AI changes the principle by allowing clear, under-
standable insights into route recommendations along with a satisfactory level of trust and
confidence.

The initiative has turned from only maximizing travel to a proactive approach to road
maintenance that allows users to report such things as potholes or damaged infrastructure.
This feedback is examined and sent to construction teams for fast remedies and improved
roads, generally. The project also significantly touches on a system of predictive mainte-
nance which successfully leverages historic and current data to allow authorities to predict
maintenance needs, cut down on maintenance costs, and prevent critical failures. The GIS
mapping, route optimization algorithms, and user-friendly interfaces epitomize a full so-
lution package-they bring meaningful improvements not only for the individual traveler
but also to the safety and resilience of a broader community.

1.1 Motivation
In the fast-paced world of today, safe and efficient transportation is a necessity, although
it is not easy because of the many unpredictable factors that make transport very hard

1
Real-Time Route Optimization with Weather XAI 2024-25
to realize. For example, an increase in bad weathering intensities and sudden closures of
road networks, together with the underdeveloped and unadvanced structures of the roads,
facilitate easy interruption of a journey and bring about delays, as well as an increased
risk, leading to commuter aggravation. Systems of conventional ways to pilot the people
have this limitation: they are not fitted to deal with actual chicanes of such time and
conditions like adaptation to altered situations or trustworthiness in solution suggesting
that is safe for the traveler.

Rising number of extreme weather scenarios due to global climate change and the emer-
gence of cities have significantly increased the need for an innovative, intelligent, and
technology-driven route optimization approach. By using real-time weather data, and ar-
tificial intelligence in order to evolve a safety and quality route, a solution might just have
been developed. It might also improve user experience by reducing disruptions. Similarly,
the incorporation of Explainable AI (XAI) in such systems is likely to add an advantage
that will boost trust and confidence through transparent decision-making.

The other significant stimulant for this project is the very fast-growing area of worry
along road infrastructure maintainability. Due to road damages like potholes and very
untidy road surfaces, there are high chances for accidents in addition to inefficiencies of
movement. Current systems lack effective ways for identifying these areas in a timely
manner and the designs for making such trials intended to improve. The project merges
end-user feedback and maintains the predictive maintenance alerts possible on integrating
the gap between the road-users and maintenance authorities by facilitating quick responses
and early interventions for controllable damages.

Ultimately, the vision for this project is to build a fully-connected, seamless transport
network built with high quality, intelligent, and streamlined operational solutions that
create better flow of products and services between the passionate customers of that fu-
ture new energy ecosystem.

1.2 Objective of the project

1.Climate-Resilient and Secure Routes:
Plan dynamic routes for users based on access to real-time weather data and road condi-
tions by having a system that ensures optimal safety and dependability in the course of

Dept.of ISE, S.I.T.,Tumakuru-03 2

Real-Time Route Optimization with Weather XAI 2024-25
travel.

2.Increase transparency with explainable AI (XAI):

Adopt Explainable AI to let the machine reveal clear and understandable insights on how
decisions about providing routes are made to promote trust in users.
3. Provide a Real-Time Method for Road Maintenance Reporting:
Road users can use this platform to report road conditions, such as potholes or damaged
infrastructure, to help ENRTF form a crucial base for action and efficient maintenance.

1.3 Organisation of the report

This report is divided into 7 chapters.
Chapter 2 presents a literature survey on the relevant studies and findings that inform this
project. It covers topics such as Explainable AI (XAI) in route optimization, real-time
road monitoring using satellite imagery, integration of weather data for disaster response,
algorithms for hazard sensing, and enhanced route optimization with user input.
Chapter 3 outlines the proposed methodology for the Real-Time Route Optimization and
Safety System. It details the data acquisition and preprocessing, route optimization sys-
tem development, and the decision-making process. The chapter also includes a system
workflow diagram and reference equations for safety score computation, route optimiza-
tion, and predictive maintenance.
Chapter 4 provides an overview of the system architecture, including weather and traffic
data integration, decentralized data processing, user interface design, and the route op-
timization process. It also describes the implementation methodology, including activity
diagrams, development and testing approaches, and system integration.
Chapter 5 discusses the software requirements and the algorithm used in the system. It
covers the programming language, frameworks, libraries, and development environment.
The chapter also explains the real-time route optimization algorithm, which includes data
acquisition, processing, scoring, and user feedback integration.
Chapter 6 presents the results of the system evaluation. It includes evaluative aspects
such as adaptability to weather conditions, efficiency of route optimization, transparency
of recommendations, scalability, and integration with road maintenance feedback. The
chapter also provides key observations, snapshots of the system interface, and a compre-

Dept.of ISE, S.I.T.,Tumakuru-03 3

Real-Time Route Optimization with Weather XAI 2024-25
hensive analysis of the project’s code metrics, API endpoint response times, component
complexity, and database schema performance.
Chapter 7 concludes the report by summarizing the project’s achievements, including the
successful implementation of climate-resilient routes, transparent decision-making using
XAI, faster road maintenance, and increased navigation efficiency. It also outlines future
development directions, such as enhanced data integration, predictive maintenance alerts,
advanced satellite imagery analysis, and dynamic traffic monitoring.

Dept.of ISE, S.I.T.,Tumakuru-03 4

Chapter 2
Literature Survey

2.1 Literature Review on Explainable AI

[2] The authors Y. Liu, R. Zhao, M. Huang, in this study explores the fusion of Ex-
plainable Artificial Intelligence (XAI) techniques with route planning systems to improve
transparency in decision-making within intelligent transportation. The authors devel-
oped a framework that applies XAI methods to interpret and explain the outputs of
AI-driven route planning algorithms. Their findings suggest that incorporating XAI into
route planning enhances decision transparency and fosters greater user trust in intelligent
transportation systems.
[5] The Authors R. Naik, D. Patel, S. Menon, in this paper introduces a hybrid model
that combines multi-modal transport optimization with Explainable AI (XAI) to enhance
decision-making transparency. The authors integrated various transportation modes—such
as buses, trains, and bicycles—into a unified optimization framework. By applying XAI
techniques, the system provides clear explanations for its recommendations, enabling users
to understand the rationale behind suggested transport combinations. Experimental re-
sults indicate improvements in both efficiency and user satisfaction.
[7] The authors M. Patel and J. H. Kim, in this paper explores the application of Gen-
erative Artificial Intelligence (AI) in optimizing delivery routes within logistics networks.
The authors developed a generative model that creates efficient routing plans by analyz-
ing vast datasets, including traffic patterns, delivery time windows, and vehicle capacities.
The model adapts to dynamic conditions, such as real-time traffic updates and weather
changes, to continuously refine route suggestions. Experimental results indicate that this
approach significantly reduces delivery times and operational costs compared to tradi-
tional routing methods.
[9]The authors P. Kumar and A. Bose, in this paper examines the role of Artificial
Intelligence (AI) in predictive route optimization within transportation systems. The

5
Real-Time Route Optimization with Weather XAI 2024-25
authors developed AI-driven models that forecast traffic conditions based on historical
data, real-time inputs, and predictive analytics. These models enable transportation
networks to anticipate congestion and adjust routing strategies accordingly. The study
reports improvements in travel time reliability and network efficiency, demonstrating the
effectiveness of AI in enhancing predictive capabilities for route optimization.

[10]The authors S. L. Roberts and K. B. Yadav,in this comprehensive review addresses

the current challenges and future trends in AI-based route optimization. The authors
discuss obstacles such as data privacy concerns, integration complexities with existing
infrastructure, and the need for real-time data processing capabilities. Emerging trends
highlighted include the adoption of Explainable AI (XAI) to enhance transparency, the
integration of Internet of Things (IoT) devices for real-time data acquisition, and the
development of more robust algorithms capable of handling dynamic and uncertain envi-
ronments. The paper provides valuable insights for researchers and practitioners aiming
to advance AI-driven route optimization solutions.

[13]The authors N. J. Sullivan and A. B. Kapoor, in this study investigates the applica-
tion of Artificial Intelligence (AI) in optimizing multi-modal transportation routes within
smart city environments. The authors developed an AI-based platform that integrates
various transportation modes—including buses, trains, bicycles, and walking paths—into
a cohesive routing system. By analyzing factors such as schedule synchronization, user
preferences, and real-time traffic data, the platform provides optimal route recommenda-
tions that enhance commuter convenience and reduce travel times. The system’s adapt-
ability to user needs and urban dynamics positions it as a valuable tool for smart city
transportation planning.

[15]In this paper, the authors M. A. Rahman and L. K. Zhao explored the use of deep
learning models for pothole detection to enhance road safety. They developed a convolu-
tional neural network (CNN)-based approach that analyzes road surface images captured
from vehicle-mounted cameras. The model efficiently detects potholes, classifies their
severity, and provides real-time alerts to drivers and road maintenance authorities. Ex-
perimental results demonstrated high detection accuracy, showcasing the potential of AI
in improving road conditions and reducing accident risks.

[16]The authors L. P. Richards and D. N. Kumar, in this paper examines AI-powered

navigation systems for autonomous vehicles in complex, dynamic environments. The au-

Dept.of ISE, S.I.T.,Tumakuru-03 6

Real-Time Route Optimization with Weather XAI 2024-25
thors proposed a multi-layer AI framework that integrates sensor fusion, deep reinforce-
ment learning, and real-time traffic analytics to enable autonomous vehicles to navigate
safely. The system continuously learns from environmental data, making adaptive driv-
ing decisions to avoid obstacles and optimize routes. Simulation results demonstrate that
AI-driven navigation enhances vehicular safety, efficiency, and adaptability in urban and
highway scenarios.

2.2 Literature Review on Route Optimisation

[1] The authors H. Wang, X. Li and J. Zhang, in this paper presents a dynamic route opti-
mization framework that integrates real-time weather data with machine learning models
to enhance transportation efficiency. The authors developed predictive models that as-
sess the impact of various weather conditions on traffic flow and safety. By incorporating
these models into route planning algorithms, the system can proactively adjust routes in
response to changing weather patterns, thereby improving travel time and safety.

[3] The authors P. Gupta, A. Verma and R. Kumar, in this paper addresses the chal-
lenge of real-time pothole detection to improve navigation safety. The authors employed
Convolutional Neural Networks (CNNs) to accurately identify potholes from live video
feeds captured by vehicle-mounted cameras. The system processes video frames in real-
time, enabling immediate detection and localization of road surface anomalies. Field
tests demonstrated the model’s effectiveness in various driving conditions, highlighting
its potential for integration into advanced driver-assistance systems.

[4] The Authors K. Sharma and S. Banerjee, in this study presents a methodology for
assessing route safety by analyzing weather patterns through satellite imagery and Geo-
graphic Information System (GIS) mapping. The authors developed a system that pro-
cesses satellite data to detect weather-related hazards, such as heavy rainfall or fog, and
maps these hazards onto transportation networks. This approach enables the identifica-
tion of potentially dangerous routes, allowing for proactive measures to ensure traveler
safety.

[6] The authors H. Nakamura and S. K. Lee, in this study proposes an extension to the
traditional Dijkstra algorithm for shortest path computation. The authors introduced
modifications that allow the algorithm to handle dynamic edge weights, accommodat-
ing real-time changes in network conditions such as traffic congestion or road closures.

Dept.of ISE, S.I.T.,Tumakuru-03 7

Real-Time Route Optimization with Weather XAI 2024-25
The extended algorithm demonstrates improved performance in dynamic environments,
making it suitable for real-time navigation systems.

[8] The authors C. Zhang and E. P. Thomas, in this paper explores the application
of Generative Artificial Intelligence (AI) in optimizing delivery routes within logistics
networks. The authors developed a generative model that creates efficient routing plans
by analyzing vast datasets, including traffic patterns, delivery time windows, and vehicle
capacities. The model adapts to dynamic conditions, such as real-time traffic updates and
weather changes, to continuously refine route suggestions. Experimental results indicate
that this approach significantly reduces delivery times and operational costs compared to
traditional routing methods.

[12] The authors M. A. Rahman and L. K. Zhao, in this paper presents a methodology for
monitoring road conditions using satellite imagery to enhance navigation systems. The
authors developed an image processing framework that analyzes high-resolution satellite
images to detect road surface anomalies, such as potholes, cracks, and debris. The ex-
tracted information is integrated into navigation systems, providing real-time updates
on road conditions to drivers and autonomous vehicles. The approach aims to improve
route planning by incorporating up-to-date information on road quality, thereby enhanc-
ing safety and comfort.

[14] The Authors B. R. Lewis and C. T. Wong, in this paper explores the integration
of Artificial Intelligence (AI) and Internet of Things (IoT) technologies for predictive
maintenance of road infrastructure. The authors developed a system where IoT sensors
embedded in roadways collect data on structural health indicators, such as vibrations and
surface wear. AI algorithms analyze this data to predict maintenance needs before critical
failures occur.

2.3 Summary
Recent advancements in integrating real-time weather data and machine learning, along-
side explainable AI (XAI), have significantly improved route optimization and travel
safety, with studies showing high accuracy in predicting optimal routes under varying
conditions. Satellite imagery and AI-driven remote sensing have enhanced road condi-
tion monitoring and infrastructure maintenance, particularly in rural areas. AI, coupled
with XAI, has also been applied in disaster risk reduction, aviation forecasting, and de-

Dept.of ISE, S.I.T.,Tumakuru-03 8

Real-Time Route Optimization with Weather XAI 2024-25
livery route optimization, improving safety, efficiency, and decision transparency. Further
developments in deep learning and geospatial AI are optimizing traffic flow, predictive
maintenance, and multimodal transport systems. Challenges like data biases and compu-
tational efficiency persist, but AI’s role in creating sustainable and resilient transportation
networks continues to expand, improving both urban mobility and disaster resilience.

Dept.of ISE, S.I.T.,Tumakuru-03 9

Chapter 3
System Requirements
This section outlines the following software and hardware tools that are used to develop
the Real-Time Route Optimization and Safety System. The selected technologies support
solid implementation and seamless integration.

3.1 Software:
This project requires a robust software stack to ensure efficient data processing, machine
learning model execution, and seamless user interaction. The system is compatible with
Windows 10/11, Ubuntu 20.04+, and macOS, providing flexibility in development and
deployment. The core programming languages used are Python, which handles AI and
machine learning computations, and JavaScript, which is used for building an interactive
web interface.
For AI and machine learning, the system leverages PyTorch for model training and infer-
ence, Scikit-learn for machine learning algorithms, and LIME for Explainable AI (XAI) to
provide transparent decision-making insights. while Geopandas and Folium help manage
geospatial data and interactive mapping. The backend is developed using Flask , ensuring
efficient handling of API requests and database interactions, while the frontend is built
using React.js providing a dynamic and responsive user experience.
For data storage and management, SQLlite is used as the primary relational database .
The system integrates third-party APIs such as WeatherAPI to fetch real-time weather
conditions and Gomaps Pro for route mapping and navigation.
With this comprehensive software stack, the system efficiently integrates AI, real-time
data, and user feedback to enhance route optimization and road safety.

3.2 Hardware
This project requires a well-equipped hardware setup to efficiently process real-time data,
execute machine learning models, and handle GIS mapping. The minimum hardware

10
Real-Time Route Optimization with Weather XAI 2024-25
requirements include an Intel Core i5 or AMD Ryzen 5 processor, 8GB of RAM, a 256GB
SSD, and a stable internet connection for fetching real-time weather. A basic integrated
GPU can support normal operations, but performance may be limited when handling
large-scale AI computations and image processing.
For optimal performance, a high-end configuration is recommended, including an Intel
Core i7/i9 or AMD Ryzen 7/9 processor, 16GB or more RAM, and a 512GB SSD or
higher to handle large datasets efficiently. A high-speed fiber-optic internet connection is
essential for real-time data retrieval and API requests.
Additional hardware may be required for real-world implementation and testing, such as
GPS modules for real-time location tracking and edge computing devices like Raspberry
Pi or NVIDIA Jetson Nano for IoT-based applications. External HDDs or cloud storage
are recommended for dataset backups and long-term data retention. If on-field weather
data collection is necessary, weather sensors can be integrated to complement API-based
weather updates.

3.3 Summary
This project is to improve route planning and road safety. Built with Python (AI/ML) and
JavaScript (frontend), it integrates tools like TensorFlow and Geopandas for data process-
ing and machine learning. The system relies on APIs like WeatherAPI and Gomapspro
Maps for real-time data and uses SQLlite for storage. For optimal performance, the
system requires a high-performance hardware setup, including an Intel Core i7 processor,
16GB+ RAM, 512GB SSD, and an NVIDIA RTX 3060 GPU. It can be deployed on cloud
platforms or on-premise servers.

Dept.of ISE, S.I.T.,Tumakuru-03 11

Chapter 4

Proposed Methodology

This chapter lays out the proposed Real-Time Route Optimization and Safety System,
defining its methodology and laying out the overall workflow of the system. The system
uses real-time weather data and XAI to bring a safer yet more efficient commute and
enhance repair operations on the roads.

4.1 Introduction about Methodology

The Real-Time Route Optimization and Safety System enhances travel safety and effi-
ciency by integrating real-time weather data with intelligent route planning. The system
uses the OpenWeather API to fetch live updates on temperature, precipitation, wind
speed, and hazardous weather conditions that could affect road travel. Alongside this,
GoMaps Pro is employed to fetch optimized routes, ensuring travelers receive the safest
and most efficient guidance. By combining these technologies, the system dynamically
analyzes road conditions and weather risks, providing adaptive route recommendations
that adjust to real-time changes.

An essential feature is the incorporation of Explainable AI (XAI), which ensures trans-

parency in the decision-making process. XAI allows users to understand why a particular
route is recommended, based on factors like weather conditions and traffic data, fostering
trust and enabling informed decisions. For example, if a route is avoided due to dangerous
weather, the system explains why, ensuring users feel confident in its recommendations.

In addition, the system acts proactively by supporting real-time road maintenance. When
hazardous conditions, such as accidents or weather-related hazards, are detected, the sys-
tem alerts authorities, helping them respond quickly to mitigate risks. This functionality
not only improves individual travel safety but also aids in the timely maintenance of road
infrastructure, enhancing overall safety and efficiency.

12
Real-Time Route Optimization with Weather XAI 2024-25

4.2 System Architecture

Figure 4.1: System Architecture

As Shown in the figure 4.1, The system architecture integrates real-time weather data,
satellite imagery, and Explainable AI (XAI) to optimize route recommendations for en-
hanced travel safety and efficiency. The user interface allows users to input their source
and destination and displays optimized routes, safety scores, and explanations through
XAI, fostering transparency and user trust.
The Data Ingestion Layer gathers real-time weather data via the OpenWeather API,
providing crucial information such as temperature, wind speed, and precipitation. The

Dept.of ISE, S.I.T.,Tumakuru-03 13

Real-Time Route Optimization with Weather XAI 2024-25
Google Maps API (GoMaps) is used for geospatial data, road networks, and traffic infor-
mation, enabling accurate route visualization.

In the Processing Analysis Layer, the YOLO model analyzes satellite imagery to detect
hazards like landslides or road cracks, while weather data processing combines forecasts
and historical data to predict unsafe conditions. The system stores and organizes the
processed data in a database for real-time access.

The Decision-Making Layer uses hazard detection systems to identify unsafe areas. Op-
timization algorithms, such as Extended Dijkstra, calculate routes that balance safety
and efficiency. The Explainable AI component (LIME) explains the reasons behind route
recommendations, enhancing user confidence.

The Output Layer delivers optimized routes with safety scores and justifications, while
the Road Repair Alert System notifies maintenance teams of critical issues like potholes
or accidents.

The architecture aligns with the project’s objectives by creating climate-resilient routes,
ensuring transparency through XAI, and enabling efficient road repairs. The system flow
begins with user input, followed by real-time data retrieval, hazard detection, route op-
timization, and dynamic updates to ensure safety. The system adapts to changing road
and weather conditions, providing a user-centric experience and contributing to more sus-
tainable transportation systems. The system also continuously monitors road conditions
and provides real-time alerts to users regarding any changes in weather or traffic that
could impact their journey. Additionally, the integration of user feedback helps refine
the system’s hazard detection and route optimization algorithms, improving the accu-
racy and reliability of future recommendations. By ensuring ongoing system updates and
maintenance, the architecture remains adaptive to new challenges and technologies. Ul-
timately, the system contributes to safer, more efficient travel while supporting proactive
infrastructure maintenance.

4.3 System Workflow

The system workflow diagram depicts the entire route optimization lifecycle, from data
acquisition to decision-making and feedback incorporation.

Dept.of ISE, S.I.T.,Tumakuru-03 14

Real-Time Route Optimization with Weather XAI 2024-25

Figure 4.2: System Workflow for the Real Time Route Optimization with Weather & XAI

Data Input: Users input source and destination, and the system retrieves real-time
weather data using the OpenWeather API.
Preprocessing: Weather data and road conditions are analyzed using the GoMaps Pro
API to identify potential hazards.
Route Computation: The algorithm computes optimized routes, avoiding unsafe regions,
and assigns a safety score.
User Notification: The recommended route is displayed along with an explanation of the

Dept.of ISE, S.I.T.,Tumakuru-03 15

Real-Time Route Optimization with Weather XAI 2024-25
safety score using LIME for transparency.
Feedback Integration: User feedback on road conditions is collected and integrated into
future recommendations.
Repair Alerts: Critical road problems are reported to the authorities in time for repair.

4.4 Implementation
The Real-Time Route Optimization and Safety System integrates GoMaps Pro for route
mapping and OpenWeather API for real-time weather data to provide intelligent and
adaptive navigation. The process begins when a user inputs their starting point and des-
tination, triggering a request to GoMaps Pro, which fetches multiple route options based
on current road conditions, traffic congestion, and estimated travel time. Simultaneously,
the system queries OpenWeather API to obtain live weather data, analyzing conditions
such as rainfall, fog, storms, and extreme temperatures along each suggested route.
Using this real-time data, the system dynamically assigns risk scores to different road
segments by evaluating weather severity and its potential impact on driving safety. Routes
with higher risk scores—due to factors like heavy rainfall reducing visibility or icy roads
increasing the likelihood of accidents—are de-prioritized in favor of safer alternatives.
The final optimized route is then selected, balancing efficiency and safety before being
presented to the user.
To enhance transparency and user confidence, the system integrates Explainable AI (XAI),
which provides justifications for the chosen route, detailing how weather conditions and
traffic patterns influenced the decision. This ensures that users not only receive the safest
and most efficient path but also understand the reasoning behind the recommendations.
By leveraging real-time data, dynamic risk assessment, and AI-driven insights, this system
offers a climate-resilient, intelligent, and user-centric navigation experience, making road
travel safer and more reliable in ever-changing conditions.

4.5 Performance Measurement

The effectiveness of a route optimization system can be evaluated based on several key
performance metrics that ensure efficiency, safety, scalability, and user trust. Time for
Route Computation is a crucial factor, representing the time taken by the system to
process user inputs, including the source and destination, and generate an optimized route
under various conditions. A highly efficient system should minimize computational delays

Dept.of ISE, S.I.T.,Tumakuru-03 16

Real-Time Route Optimization with Weather XAI 2024-25
while considering factors such as real-time traffic, road conditions, and weather updates.
Another critical aspect is the Route Safety Index, which assesses the system’s ability to
recommend routes with lower safety risks by factoring in dynamic elements such as adverse
weather conditions, road closures, traffic congestion, and accident-prone areas. A well-
designed system should prioritize user safety by continuously updating risk assessments
based on real-time data.
Additionally, the Scalability of the System is analyzed by evaluating its performance
under high workloads, such as processing large volumes of real-time weather data and
handling multiple user requests simultaneously. A robust system must maintain effi-
ciency and accuracy even when faced with high traffic demands. The Responsiveness
of Feedback measures how effectively the system utilizes user-reported hazards, such
as potholes, roadblocks, or accidents, and communicates this information to road main-
tenance teams for quick resolution. A rapid feedback loop ensures that road conditions
are promptly updated, improving overall navigation reliability. Finally, the User Trust
Score assesses how well the system’s explainable AI (XAI) features enhance user con-
fidence in the recommended routes. By analyzing user comprehension and trust in the
provided explanations, developers can refine the system to ensure greater transparency
and user satisfaction. Together, these metrics contribute to a comprehensive evaluation
of the system’s performance, reliability, and effectiveness in delivering safe and optimized
navigation solutions.

Dept.of ISE, S.I.T.,Tumakuru-03 17

Chapter 5
Results
This chapter provides an objective evaluation of the Real-Time Route Optimization and
Safety System Using Weather Data and Explainable AI. The findings demonstrate the
system’s performance, adaptability, and reliability from different evaluation perspectives.
The main considerations in the evaluation encompass the system’s capacity for reacting
to the unpredictable nature of weather changes, delivery of the optimized routes, effective
use of Explainability in XAI, and the ability to take into account real-time operational
handling.

5.1 Snapshots

Figure 5.1: User Interface

The RouteX AI interface is a web-based Intelligent Route Optimization System designed

for weather-aware navigation. It provides real-time route planning by analyzing weather,
traffic, and road conditions to ensure safety and efficiency. Key features include Plan
Route, Safety Check, Live Navigation, Road Feedback, and Pothole Detection, each with
real-time updates and user feedback integration. The system displays operational status

18
Real-Time Route Optimization with Weather XAI 2024-25
and regularly updates weather data for accurate routing. The UI is clean, modern, and
user-friendly, ensuring transparency and ease of use.

Figure 5.2: Road Condition Reports

The Road Condition Reports interface offers real-time updates on various road hazards
such as potholes, cracks, and waterlogging. Each report provides critical details, including
the issue type (e.g., pothole, crack, or waterlogging), the severity level (e.g., low, medium),
and the current status of the issue (e.g., pending). The interface also displays the location,
including GPS coordinates and place names, along with a detailed description of the
problem. Users can upload images as evidence, and each report is timestamped to indicate
when the issue was reported. This system helps in tracking and addressing road conditions
efficiently.
This system enables users to report and track road conditions for safer navigation. The
pending status suggests that reports are awaiting verification or action. The interface is
organized, user-friendly, and prioritizes transparency in road safety.

Figure 5.3: Route Explorer

Dept.of ISE, S.I.T.,Tumakuru-03 19

Real-Time Route Optimization with Weather XAI 2024-25
The Route Explorer interface facilitates route planning and provides real-time weather
insights for journeys, such as the one from Tumkur to Sakleshpur. It offers key travel
details, including the duration (3 hours 17 minutes) and distance (190 km). The weather
insights feature provides real-time conditions at key points along the route, such as at the
midpoint (Tiptur), where the weather is clear with 18°C, 66% humidity, and a wind speed
of 3.62 m/s, and at the destination (Sakleshpur), where there are few clouds with 16°C,
58% humidity, and a wind speed of 1.69 m/s. An interactive map displays the optimized
route with waypoints, making trip planning more efficient by integrating real-time weather
data, enabling users to make well-informed travel decisions.

This system enhances trip planning by integrating route optimization with real-time
weather data, helping users make informed decisions.

Figure 5.4: Road Condition Feedback Form

The Road Condition Feedback interface allows users to report road issues by providing key
details: The location of the issue is Chikkanayakanahalli, with the detected coordinates
being 13.3264, 77.1275. The road condition is identified as having potholes, with a severity
level classified as medium, indicating a moderate problem. The description provided for
the condition is “Full of potholes.” Users are given the option to upload an image as
evidence of the issue. A “Submit Feedback” button is available, allowing users to report
the problem and contribute valuable data for system improvement.

This system enables real-time road condition reporting, helping authorities or navigation
systems improve road safety and maintenance.

Dept.of ISE, S.I.T.,Tumakuru-03 20

Real-Time Route Optimization with Weather XAI 2024-25

Figure 5.5: Pothole Detection

Figure 5.6: Live Navigation

The Live Navigation system provides real-time route guidance with safety alerts. This
system ensures efficient route planning while providing real-time weather and traffic alerts
for safer travel.

5.2 Comprehensive Analysis of Route XAI Project

5.2.1 Code Metrics Analysis

Dept.of ISE, S.I.T.,Tumakuru-03 21

Real-Time Route Optimization with Weather XAI 2024-25

File Name Size Functionality Description

(Bytes)
app.py 19,113 Primary Flask application containing route
handlers, image processing logic, and API
endpoints. Contains 15+ route handlers and
core business logic.
safe route planner.py 27,225 Implements the route planning algorithm
with safety considerations. Largest single file
indicating complex route optimization logic.
models.py 1,161 Database schema definitions using
SQLAlchemy ORM. Relatively small
size indicates clean data model design.

Table 5.1: File Size Distribution and Functionality Analysis

Dept.of ISE, S.I.T.,Tumakuru-03 22

Chapter 6
Conclusion
The project, Real-Time Route Optimization and Safety System Using Weather Data and
Explainable AI, has been successfully implemented and rigorously tested, proving to be
highly effective in addressing critical challenges related to navigation and road safety. By
leveraging real-time weather data, GIS mapping, satellite imagery, and Explainable AI
(XAI), the system is capable of offering the most efficient and safest routes for users while
ensuring full transparency in the decision-making process. Rigorous testing in diverse sce-
narios has demonstrated the system’s robustness, adaptability, and its ability to reduce
travel risks significantly while enhancing road safety. The system also provides a com-
prehensive view of potential hazards, ensuring users are always informed and prepared
during their travels.
The successful implementation of the project has achieved several core objectives:
The implementation successfully achieves its core objectives by:

• Climate-Resilient Routes: The system intelligently identifies and suggests travel

paths that avoid areas with hazardous weather conditions, such as heavy rainfall,
fog, or snow, ensuring that users are always traveling on safer routes. Additionally,
it dynamically considers real-time road closures, accidents, or obstacles, providing
the highest safety score for each recommended route. By factoring in these variables,
the system offers routes that reduce exposure to unforeseen risks.

• Transparent Decision-Making: The system intelligently identifies and suggests

travel paths that avoid areas with hazardous weather conditions, such as heavy
rainfall, fog, or snow, ensuring that users are always traveling on safer routes. Ad-
ditionally, it dynamically considers real-time road closures, accidents, or obstacles,
providing the highest safety score for each recommended route. By factoring in
these variables, the system offers routes that reduce exposure to unforeseen risks.

• Faster Road Maintenance: Provides valuable feedback to road maintenance

23
Real-Time Route Optimization with Weather XAI 2024-25
teams, allowing them to act more quickly on priority issues related to road mainte-
nance.

• Increased Navigation Efficiency: Reduces travel time, as well as risks associ-

ated with journeys made in changing conditions that may include poor weather or
unforeseeable road hazards; smoother travel overall.

Integrating machine learning algorithms, real-time weather monitoring, GIS mapping,

and satellite imagery, we created a strong, scalable, and user-centric framework for safe
and efficient navigation. XAI took on the role of informing users not only about the
system’s reasoning behind its recommendation but also about how to trust it. In addition,
the modular architecture of the system supports seamless upgrades and enhances future
development.

Dept.of ISE, S.I.T.,Tumakuru-03 24

Chapter 7
Future Work
1. Predictive Maintenance: It alerts AI-driven models to analyze historical and
real-time road condition data, predicting when and where repairs might be needed.
By collecting data from sources such as satellite imagery, IoT sensors embedded in
roads, vehicle telematics, and crowdsourced reports, the system ensures comprehen-
sive road monitoring. Machine learning models process this data, identifying road
degradation patterns influenced by weather conditions and traffic load to estimate
potential damage at specific locations. Additionally, real-time analysis integrates
data from connected vehicles, traffic cameras, and road sensors to detect early signs
of deterioration, such as cracks, vibrations, and potholes. An automated alert sys-
tem prioritizes repair needs and notifies municipal authorities, enabling quick and
strategic maintenance planning. This proactive approach results in faster identi-
fication of road issues, reduced repair costs, and improved road safety, ultimately
ensuring smoother traffic flow and enhanced transportation efficiency.

2. Collaboration with authorities: It involves the development of an automated

notification system that directly alerts road maintenance teams and local govern-
ment agencies about reported road issues. A cloud-based platform collects real-
time reports from AI-driven road monitoring systems, sensors, and user feedback,
ensuring comprehensive data aggregation. To enhance efficiency, an AI-driven clas-
sification system categorizes reports based on severity, potential risk, and traffic
density, allowing authorities to prioritize urgent repairs. Seamless integration with
municipal databases and public works departments enables real-time tracking and
streamlined communication between stakeholders. Additionally, a dedicated mobile
app provides maintenance teams with a dashboard to view reports, monitor repair
progress, and receive instant alerts about critical road conditions. This automated
approach leads to faster response times, reduced traffic congestion due to prompt

25
Real-Time Route Optimization with Weather XAI 2024-25
repairs, and improved coordination between public and private entities, ultimately
enhancing road safety and infrastructure management.

3. Gamification: It enhances user engagement by incorporating a reward-based sys-

tem that encourages active participation in reporting road conditions. A user-
friendly mobile app allows individuals to report road issues, traffic conditions, and
hazards using GPS tagging, ensuring accurate location-based data collection. To
motivate consistent contributions, a points and badges system rewards users for sub-
mitting frequent and precise reports, verifying existing data, and sharing updates.
Additionally, leaderboards, weekly challenges, and milestone-based rewards create
a competitive and engaging experience, further incentivizing participation. Collab-
orations with local businesses, ride-hailing services, and fuel stations can provide
real-world incentives such as discounts, vouchers, or free services for top contrib-
utors. This approach increases user participation, improves the accuracy of road
condition data, and fosters a strong sense of community involvement in road safety
initiatives, ultimately leading to more efficient infrastructure management.

4. Multi-modal integration: It enhances route optimization by incorporating al-

ternative transportation options, ensuring accessibility for all commuters. An AI-
powered route planning system intelligently suggests the best mix of transportation
modes based on user preferences, real-time traffic, and weather conditions, optimiz-
ing travel efficiency. By integrating public transport options such as buses, metros,
trains, and ride-sharing services, users can seamlessly switch between different modes
for a more convenient journey. Additionally, pedestrian-friendly and cyclist-friendly
routes are included, highlighting safe roads, bike lanes, and weather conditions that
may impact non-motorized travel. To ensure inclusivity, the system also supports
users with disabilities by factoring in wheelchair-accessible routes, elevator avail-
ability, and curb cuts, making urban mobility more accessible. This comprehensive
approach leads to improved traffic distribution, a reduced carbon footprint, and en-
hanced transportation accessibility, fostering a more sustainable and efficient urban
commuting experience.

Dept.of ISE, S.I.T.,Tumakuru-03 26

Bibliography
[1] H. Wang, X. Li and J. Zhang, “Dynamic Route Optimization Considering Real-time
Weather Data and Machine Learning Models”, Journal of Intelligent Transportation
Systems, vol. 28, no. 2, pp. 150–165, Feb. 2023.

[2] Y. Liu, R. Zhao, M. Huang,“Integration of Explainable Artificial Intelligence with

Route Planning for Enhanced Decision Transparency” , IEEE Transactions on In-
telligent Transportation Systems, vol. 24, no. 6, pp. 4401–4412, June 2022, doi:
10.1109/TITS.2022.3176351.

[3] P. Gupta, A. Verma and R. Kumar, “Real-Time Navigation with Pothole Detection
Using Convolutional Neural Networks”, Proceedings of the 2023 IEEE International
Conference on Intelligent Systems and Applications, pp. 223–229, Nov. 2023

[4] K. Sharma and S. Banerjee, “Weather-Driven Route Safety Assessment Using Satel-
lite Imagery and GIS Mapping”, International Journal of Advanced Computational
Science and Engineering, vol. 12, no. 4, pp. 370–383, Oct. 2022.

[5] R. Naik, D. Patel, S. Menon, “A Hybrid Approach Toward Multi-Modal Transport

Optimization with Explainable AI Models”, International Journal of Data Science and
Artificial Intelligence, vol. 8, no. 3, pp. 257–268, Apr. 2024.

[6] H. Nakamura and S. K. Lee, “A Method for the Shortest Path Search by Extended
Dijkstra Algorithm”, Journal of Computational Science, vol. 36, pp. 245–260, Dec.
2019.

[7] M. Patel and J. H. Kim, “A Comprehensive Study on Optimizing Delivery Routes

through Generative AI”, Journal of Smart Logistics, vol. 29, no. 4, pp. 310–325, Oct.
2023.

[8] V. Sharma and D. H. Wilson, “Navigational Intelligence for Accident Prevention and
Real-Time Road Safety”, Transportation Research Part C: Emerging Technologies,
vol. 110, pp. 85–100, Apr. 2020.

27
Real-Time Route Optimization with Weather XAI 2024-25
[9] P. Kumar and A. Bose, “AI in Transportation: Predictive Route Optimization”, Jour-
nal of Artificial Intelligence in Transportation, vol. 18, no. 2, pp. 58–75, Aug. 2021.

[10] S. L. Roberts and K. B. Yadav, “AI Route Optimization: Challenges and Future
Trends”, Expert Systems with Applications, vol. 214, pp. 117865, Feb. 2024.

[11] C. Zhang and E. P. Thomas, “Machine Learning Applications in Dynamic Route

Planning and Traffic Prediction”, Journal of Machine Learning Research, vol. 23, pp.
1–20, Nov. 2022.

[12] R. K. Mohan and T. S. Gupta, “Satellite-Based Road Condition Monitoring for

Improved Navigation Systems”, Remote Sensing of Environment, vol. 251, pp. 112053,
Jun. 2021.

[13] N. J. Sullivan and A. B. Kapoor, “Leveraging AI for Multi-Modal Route Optimization

in Smart Cities”, Journal of Urban Computing, vol. 45, no. 1, pp. 100–118, Sep. 2023.

[14] B. R. Lewis and C. T. Wong, “Predictive Maintenance for Road Infrastructure Using
AI and IoT”, IEEE Internet of Things Journal, vol. 8, no. 5, pp. 4560–4572, May 2022.

[15] M. A. Rahman and L. K. Zhao, “Deep Learning-Based Pothole Detection for Smart
Road Safety Applications”, Journal of Intelligent Transportation Systems, vol. 34, no.
2, pp. 180–195, Jan. 2020.

[16] L. P. Richards and D. N. Kumar, “AI-Powered Navigation Systems for Autonomous

Vehicles in Dynamic Environments,” Journal of Autonomous Systems, vol. 12, no. 6,
pp. 512–530, Feb. 2024.

Dept.of ISE, S.I.T.,Tumakuru-03 28

Self- Assessment of the Project :
Level
Poor 1 Good 2 Excellent 3
PO PSO Contribution from the project
Level
1 Engineering Knowledge: Apply principles of mathematics, 3
Knowledge of mathematics, engineering data science, GIS, and AI/ML to
fundamentals engineering specialization to form of solve complex problems in route
complex engineering problems optimization and safety. The project
applies engineering principles such
as real-time weather data analysis,
satellite imagery processing,
machine learning algorithms (e.g.,
Extended Dijkstra), and Explainable
AI (XAI) to develop a dynamic route
optimization system that prioritizes
safety and efficiency.

2 Problem Analysis: Identify, formulate, review Identify inefficiencies in traditional 3

research literature and analyze complex navigation systems and formulate
engineering problems reaching substantiated AI-driven solutions. We analyzed
conclusions with consideration for sustainable limitations in existing navigation
development. systems (e.g., poor real-time
adaptation to weather/road
closures) and designed a system
integrating multi-source data
(weather, satellite imagery) and XAI
for safer, adaptive routing.

3 Design/development of solutions: Design creative Design solutions for real-time route 3

solutions for complex engineering problems and optimization with transparency and
design/develop systems/components/processes to scalability. The system integrates
meet identified needs with consideration for the GIS mapping, real-time weather
public health and safety, whole-life cost, net zero APIs, satellite imagery analysis, and
carbon, culture, society and environment as XAI models to provide climate-
required resilient routes. A user interface
displays optimized paths with
safety scores and explanations.
4 Conduct investigations of complex problems: Investigate route optimization 3
Conduct investigations of complex engineering challenges using research-based
problems using research-based knowledge methods. Extensive research
including design of experiments, modelling, evaluated existing route algorithms,
analysis & interpretation of data to provide valid weather data integration
conclusions. challenges, and XAI frameworks.
We implemented a hybrid model
combining Extended Dijkstra’s
algorithm with ML for dynamic
decision-making.

5 Modern tool usage: Create, select and apply Use AI/ML libraries, GIS tools, and 3
appropriate techniques, resources and modern XAI frameworks. We utilized
engineering & IT tools, including prediction and Python, TensorFlow, Scikit-learn,
modelling recognizing their limitations to solve OpenCV (for satellite imagery), XAI
complex engineering problems. tools (LIME/SHAP), and GIS APIs
(Google Maps) to build a scalable
and interpretable system.

6 The Engineer and the world: Address societal impact via safer 3
Analyze and evaluate societal and environmental routes and efficient road
aspects while solving complex engineering maintenance. The project enhances
problems for its impact on sustainability with road safety by avoiding hazardous
reference to economy, health, safety, legal conditions, reduces travel
framework, culture and environment. delays/fuel consumption, and
provides critical data to road repair
teams for timely maintenance,
promoting public safety and
sustainability.

7 Ethics: Apply ethical principles and commit to Ensure transparency, data privacy, 3
professional ethics, human values, diversity and and fairness in AI decisions. XAI
inclusion; adhere to national & international laws. provides clear reasoning for route
recommendations, ensuring user
trust. Location data is anonymized,
and the system avoids biases in
route suggestions.
8 Individual and Team Work: Function effectively as Collaborate in roles like data 3
an individual, and as a member or leader in processing, AI modeling, UI design,
diverse/multi-disciplinary teams. and integration. Roles were divided
among team members: weather
data integration, satellite imagery
analysis, algorithm development,
XAI implementation, and UI design,
ensuring effective collaboration.

9 Communication:Communicate effectively and Document and present technical 3

inclusively within the engineering community and and non-technical aspects. We
society at large, such as being able to comprehend created project reports, user
and write effective reports and design manuals, and presentations
documentation, make effective presentations explaining system functionality, XAI
considering cultural, language, and learning outputs, and benefits to
differences
stakeholders. Demo videos and
dashboards were developed for
clarity.

10 Project Management and Finance: Apply Adapt to evolving AI, GIS, and real- 2
knowledge and understanding of engineering time data technologies. The project
management principles and economic decision- deepened our expertise in AI/ML,
making and apply these to one’s own work, as a real-time systems, and geospatial
member and leader in a team, and to manage analysis. It encourages continuous
projects and in multidisciplinary environments.. learning as XAI and satellite imagery
technologies advance.

11 Life-long Learning: Recognize the need for, and As project is about the ongoing and 3
have the preparation and ability for i) independent upcoming technologies, further the
and life-long learning ii) adaptability to new and medical image processing is applied
emerging technologies and iii) critical thinking in to detect the tumors, diseases and
the broadest context of technological change recent covid affected lungs severity
could also be found using clustering
algorithms.
Self learning ability improved
12 PSO1: Apply the concepts of electronic circuits and Apply data science and AI concepts 3
systems to analyses and design systems related to to solve real-world transportation
Microelectronics, Communication, Signal processing problems. We applied AI/ML
and Embedded systems for solving real world algorithms, GIS mapping, and real-
problems time data processing to design a
route optimization system that
 addresses safety and efficiency
challenges in transportation.

13 PSO2: To identify problems in the area of Identify transportation 3

communication and embedded systems and provide inefficiencies and provide AI-driven
efficient solutions using modern tools/algorithms solutions. The project solves route
working in a team optimization challenges by
integrating satellite imagery,
weather data, and XAI, offering a
modern tool for safer navigation
and proactive road maintenance.
Real-Time Route Optimization with Weather XAI 2024-25

SDG Number SDG Name Level (1-3)

SDG 1 No Poverty -
SDG 2 Zero Hunger -
SDG 3 Good Health and Well-being 3
SDG 4 Quality Education 3
SDG 5 Gender Equality -
SDG 6 Clean Water and Sanitation -
SDG 7 Affordable and Clean Energy -
SDG 8 Decent Work and Economic Growth 3
SDG 9 Industry, Innovation and Infrastructure 3
SDG 10 Reduced Inequalities -
SDG 11 Sustainable Cities and Communities -
SDG 12 Responsible Consumption and Production -
SDG 13 Climate Action -
SDG 14 Life Below Water -
SDG 15 Life on Land -
SDG 16 Peace, Justice and Strong Institutions -
SDG 17 Partnerships for the Goals -

Dept.of ISE, S.I.T.,Tumakuru-03 33