Software Problem Statements

Domain: Artificial Intelligence

[Link].: 1

Problem Statement Title: Design and develop an Al/ML-based Decision Support System
for optimizing rake formation strategies for SAIL, with the initial implementation focused on
movements from Bokaro Steel Plant to CMO stockyards and customer locations.

Background
Humans Lives in a multifarious environment of audio signals that include speech and
non-speech sounds. The ability to accurately discern, interpret, and integrate these speech
and non-speech audio elements, along with a profound comprehension of the relationship
between them, constitute fundamental cognitive capability of humans. However, most
existing machine learning models recognize either speech or audio events. Further while
being strong in audio and speech perception, these models possess limited reasoning and
understanding capabilities. This motivate to build ALM Model which will be able to
simultaneously recognize and jointly understand speech and audio together with reasoning.

Description
The ALM (Audio Language Model) is an audio understanding problem statement describes a
situation where there’s a need to automatically extract meaningful information or context
from audio. The core challenge involves overcoming limitations such as background noise,
overlapping speech, speaker variation, and the complexity of identifying specific elements
like dialogue, music, sound effect, local context of language spoken reason wise and cultural
contextual meaning.

What Audio understanding entails:

Speech recognition: Converting Spoken language into text, good quality of solution available
for this sub task independently.
Non-Speech Audio Understanding: Identifying sounds other than speech such as music,
alarms, or environmental noises etc.
Speaker Diarization: Differentiating between different speakers in an audio recording.
Paralinguistic: Recognizing aspects of speech like emotion, tone, and hesitation that are not
part of the words themselves.
Audio Event Detection: Classifying specific audio events, such as a car honking, a dog
barking, aircraft sound, Fighter jet sound, or other industrial machine sound.
By Solving all these task together model can understand spoken text, paralinguistic, and
non-speech audio events simultaneously to answer complex questions about an entire audio
scene?

Dataset used for model training should belongs to Asian Regions (Mandarin, Urdu, Hindi,
Telgu, Tamil, Bangla etc.) along with English language.
Expected Solution

We aim to build a training datasets to provide joint speech and non-speech supervision.
Unfortunately, there is no existing datasets that meets our needs. The closest one is the
OpenAQA dataset which is an audio question answering datasets. Second expectation is a
AI Model trained on this generated dataset.

[Link].: 2

Problem Statement Title: Development of map-matching algorithm using AI-ML techniques

to distinguish vehicular movement on highway and service road

Description: Algorithm development using AI-ML techniques to distinguish vehicular

movement on highway and service road.

Desired Outcome: The developed algorithm shall provide the following: 1. Use the coarse
GNSS position of the moving vehicle and plot the movement on a map 2. Using
map-matching techniques, identify the movement of vehicle on the highway or the service
road 3. Apply AI-ML techniques to improvement map-matching performance.

[Link].: 3

Problem Statement Title: Design and build a multimodel Retrieval-Augmented Generation

(RAG) system leveraging a Large Language Model (LLM) for OFFLINE mode that can
ingest, index, and query diverse data formats such as DOC, PDF, Images and voice
recordings within a unified semantic retrieval framework

Background

Centre routinely handles diverse data types: PDF, DOC, Images, screenshots, recorded
calls, and free-form notes. Traditional search tools struggle with cross‑format understanding,
often isolating text, image, and audio searches. Retrieval‑Augmented Generation (RAG) can
address this by retrieving relevant data and grounding LLM outputs thus enhancing
accuracy, reducing effort & time, and enabling content transparency. Multimodal RAG can
extend this fusion to include images & audio, thus creating richer, context-aware intelligence.

Description

Participants are to build a multimodal RAG-based system that:

Ingests multimodal inputs – Extract textual content from DOCX/PDF, generate embeddings
for images, and perform speech-to-text conversion for audio.

Indexes all modalities in a shared vector space for seamless semantic retrieval.

Supports natural-language queries, retrieving relevant text snippets, images, and audio
segments.

Generates grounded summaries or answers, integrating retrieved context via LLM.

Establishes cross-format links, e.g., connecting an audio transcript segment to a cited
paragraph and screenshot.

Expected Solution

Unified Query Interface

● A simple chat or search box where users ‘type’ questions in plain language (e.g. “Show me
the report that has a description about international development in 2024 OR show the
report that references the screenshot taken at 14:32”).

● Optional - Support for different input modalities: other than text input - upload DOCX/PDF,
drag‑and‑drop images, attach audio files, or speak their query.

Semantic & Cross‑Modal Search

● Text-to-image search: Type a query like “email screenshot” to retrieve relevant images
alongside matching text passages.

● Image-to-text search: Upload or select an image (e.g. screenshot) and surface relevant
documents or audio transcript snippets that semantically match it.

● Optional - Audio-to-others: play an audio clip, then retrieve related text/ documents/
images that match what was spoken.

Citation Transparency & Source Navigation

● Every answer includes numbered citations linking back to source files.

● Users can expand citations to open the original document, view full transcript segments, or
inspect image metadata.

[Link].: 4

Problem Statement Title: Face Recognition Surveillance System and Communication

Systems for Missing Persons or Items at Simhastha Ujjain for the Police Department

Description: Simhastha Ujjain, one of the largest religious gatherings in India, witnesses
millions of pilgrims and tourists congregating in a single location. Managing such a vast
crowd poses significant challenges for the police department, particularly in ensuring the
safety and security of individuals and locating missing persons or items. The deployment of
advanced technologies like face recognition surveillance systems and efficient
communication systems is critical to address these challenges. During Simhastha, the
number of reports for missing persons and items surges, making it difficult for the police to
manage and respond promptly. Identifying missing persons in a vast, densely populated
area is challenging without advanced technology. Traditional methods of identifying missing
persons through manual verification of photos and descriptions are time-consuming and
prone to errors, leading to delayed response times and reduced chances of quickly locating
missing persons or items. Communication between different units and officers is often
delayed, affecting coordination and timely response, and the absence of a centralized
system to disseminate information about missing persons or items hampers effective
communication and coordination. Managing and monitoring large crowds to prevent and
respond to incidents is also challenging without real-time surveillance and communication,
and the risk of criminal activities increases in such large gatherings, necessitating advanced
surveillance for timely threat detection and intervention. The inability to quickly and
accurately identify and locate missing persons or items results in reduced operational
efficiency and effectiveness. Delays in responding to reports of missing persons or items
pose significant safety risks, especially for vulnerable individuals such as children or the
elderly. The police department faces an increased operational burden due to the high
volume of cases and the need for extensive manual effort, and inefficient handling of such
cases can erode public trust in the department’s ability to ensure safety and security during
large events. To overcome these challenges, deploying a face recognition surveillance
system that can identify and track missing persons in real time from surveillance footage,
integrating it with existing CCTV infrastructure, and implementing automated alerts for
immediate officer notification are essential. Establishing a centralized communication
platform for real-time coordination among police units, providing mobile connectivity for
officers on the ground, and enabling data sharing with the public can ensure faster and more
efficient information exchange. Comprehensive training for officers on using these
technologies and continuous technical support for system maintenance and troubleshooting
are crucial. Additionally, robust data privacy measures and strict security protocols must be
implemented to protect personal information and prevent unauthorized access. By
addressing these challenges through the implementation of advanced face recognition
surveillance and enhanced communication systems, the police department can significantly
improve its ability to manage large crowds, ensure public safety, and respond promptly to
reports of missing persons or items during Simhastha Ujjain.

[Link].: 5

Problem Statement Title: Development and Implementation of Face Recognition

Technology in the Police Department

Description: The police department aims to enhance its operational capabilities through the
development and implementation of face recognition technology (FRT). This advanced
technology can significantly aid in identifying suspects, locating missing persons, and
maintaining public safety. However, the deployment of FRT involves complex challenges that
need to be addressed to ensure its effectiveness, reliability, and ethical use.

Issues Faced 1. Technical Challenges: - Accuracy and Reliability: Developing FRT that
provides consistent and accurate results across various conditions (lighting, angles,
occlusions) and diverse populations is technically challenging. False positives and negatives
can undermine the credibility and utility of the system. - Integration with Existing Systems:
The new technology must seamlessly integrate with existing police databases and IT
infrastructure, which can be complex and resource-intensive. 2. Data Quality and
Management: - Database Quality: The effectiveness of FRT is heavily dependent on the
quality and completeness of the reference database. Inaccurate or outdated data can lead to
misidentification. - Data Volume and Speed: Handling large volumes of data in real-time
requires robust data processing capabilities and efficient algorithms. 3. Ethical and Privacy
Concerns: - Privacy Invasion: FRT has the potential to infringe on individuals' privacy rights,
leading to public distrust and resistance. There is a need to balance surveillance benefits
with privacy protections. - Bias and Discrimination: FRT systems can exhibit bias against
certain demographic groups, leading to unfair treatment and potential civil rights violations.
Ensuring fairness and mitigating bias are crucial. 4. Legal and Regulatory Issues: - Lack of
Clear Regulations: The rapid advancement of FRT has outpaced the development of
comprehensive legal frameworks. Clear regulations are needed to govern its use and protect
citizens' rights. - Compliance with Existing Laws: The development and use of FRT must
comply with existing privacy and data protection laws, which may vary across jurisdictions. 5.
Operational and Training Needs: - User Training: Effective use of FRT requires training law
enforcement personnel in its operation, interpretation of results, and ethical considerations. -
Maintenance and Support: Ongoing technical support and system maintenance are
necessary to ensure the technology remains effective and up-to-date. 6. Public Perception
and Acceptance: - Transparency and Accountability: Building public trust requires
transparency in how FRT is used and accountability mechanisms to address misuse. -
Community Engagement: Engaging with the community to educate them about the benefits
and safeguards of FRT can help mitigate concerns and gain public support. Impact on
Operations 1. Enhanced Identification Capabilities: When effectively implemented, FRT can
significantly enhance the ability to identify and apprehend suspects, improving public safety.
2. Resource Allocation: Development and implementation require substantial financial and
human resources, which must be justified by the benefits and outcomes. 3. Legal and Ethical
Implications: Misuse or ineffective use of FRT can lead to legal challenges, ethical dilemmas,
and public backlash. 4. Operational Efficiency: Proper integration and use of FRT can
improve operational efficiency, reducing the time and effort required for manual identification
processes. Solution Requirements 1. Development of Robust Algorithms: Focus on creating
advanced algorithms that ensure high accuracy, minimize bias, and handle diverse
conditions effectively. 2. Comprehensive Legal Framework: Advocate for and contribute to
the development of clear regulations that govern the ethical use of FRT. 3. Privacy and
Security Measures: Implement strong data protection protocols to safeguard biometric data
and ensure compliance with privacy laws. 4. Training Programs: Develop comprehensive
training programs for law enforcement personnel to ensure they can use FRT effectively and
ethically. 5. Public Engagement Initiatives: Establish initiatives to engage with the
community, address concerns, and build public trust in the use of FRT. 6. Continuous
Monitoring and Evaluation: Set up mechanisms for continuous monitoring, evaluation, and
improvement of the FRT system to ensure it meets operational needs and ethical standards.
By addressing these challenges, the police department can develop and implement face
recognition technology that enhances law enforcement capabilities while maintaining public
trust, protecting privacy, and ensuring ethical use.

[Link].: 6

Problem Statement Title: Development of AI/ML based solution for detection of face-swap
based deep fake videos

Description 1: Synthetically-generated audios and videos i.e. deep fakes are in trends
nowadays. Although these have caught the imagination of the computer and tech-savvy
generation, at the same time, these have created concerns because of their ability to disrupt
a nation’s politics, commit frauds, create disinformation, and produce non-consensual
content. The use of deep fakes has also been seen recently to purposefully defame the
character of a person. Due to enhancements in AI/ML and Large Language Models for
Generative AI, identification and detection of deep fakes have created a huge challenge for
Security Agencies. Research on multiple aspects of deep fake identification has been
started. As a use case for the problem statement, a forensic technique to authenticate
face-swap deep fake videos in which a person's facial identity is replaced with another's may
be developed.

Description 2: Addressing the deep fake problem requires a multi-pronged approach,

combining technology, regulation, education, and collaboration to mitigate the risks and
protect individuals and society. However, on the technological front, detecting deep fakes
involves development and testing with numerous advanced algorithms and tools. Here are
some of the prominent approaches and technologies that can be used in deep fake
detection:

i. Convolutional Neural Networks (CNNs): CNNs can be trained to detect inconsistencies in

facial features, expressions, and movements. CNNs can also analyze video frames over
time to identify unnatural transitions and discrepancies.

ii. Recurrent Neural Networks (RNNs) and Long Short-Term Memory (LSTM) Networks:
RNNs and LSTMs can be used to analyze the sequence of frames in a video to detect
temporal anomalies and inconsistencies indicative of deep fakes.

iii. Capsule Networks: Capsule networks can identify discrepancies in facial pose and
texture.

iv. Adversarial Training: Adversarial training involves using Generative Adversarial Networks
(GANs) to generate deep fakes and simultaneously training another model to detect them,
improving the robustness of detection algorithms.

v. Audio-Visual Inconsistencies: Combining audio and visual analysis can also help detect
deep fakes by identifying mismatches between lip movements and speech or
inconsistencies in ambient sounds.

vi. Blockchain Technology: Using blockchain to create immutable records of media, providing
a verifiable chain of custody for digital content to ensure its authenticity.

vii. Frequency Analysis: Analyzing the frequency domain of images and videos to detect
anomalies and artifacts introduced during the deep fake creation process.

viii. Biometric Verification: Analyzing subtle behavioral traits such as micro-expressions, eye,
and head movement to detect anomalies.

ix. Hybrid Models: Using a combination of spatial, temporal, audio, and frequency analysis to
improve the robustness and accuracy of deep fake detection.

Expected solution: A solution based on AI/ML technology may be developed to authenticate

face-swap deep fake videos in which a person's facial identity is replaced with another's.
This technique may be based on any of the above-mentioned methods or it may combine
multiple methods and demonstrate the final deliverable of detection of deep fakes with
accuracy. The input of the system will be a video that is suspected of being a deep fake, and
the output should contain a detailed report with respect to various characteristics of deep
fake e.g. whether confirmed fake (details of abnormalities observed), and mathematical
underlying techniques of its creation.

[Link].: 7

Problem Statement Title: AI-Driven Crop Disease Prediction and Management System

Description 1: Crop diseases can devastate yields, leading to significant financial losses for
farmers. Early detection and timely intervention are crucial for effective management.
Description 2: Develop an AI-driven system that analyzes crop images and environmental
data to predict potential disease outbreaks. This system will provide farmers with actionable
insights and treatment recommendations to mitigate risks.

Expected Solution: A mobile and web-based application that utilizes machine learning
algorithms to identify crop diseases and suggest preventive measures and treatments based
on real-time data.

Domain: Cloud Computing

[Link].: 8

Problem Statement Title: An online system to automatically verify new title submissions by
checking for similarities with existing titles

Description 1: Press Registrar General of India (PRGI) maintains a database containing

approximately 160,000 titles. When a user submits a new title for verification, we need to
check its similarity against the existing titles in our database. The goal is to ensure that the
new title does not duplicate or closely resemble any existing title to avoid confusion and
maintain uniqueness. Additionally, the system must enforce specific guidelines to ensure that
certain words are not used, combinations of existing titles are not allowed, and titles with
similar meanings or periodicity modifications are rejected.

Description 2: Develop a system to automatically verify new title submissions by checking for
similarities with existing titles in the database and ensuring compliance with specific
guidelines. The system should reject titles that are too similar to existing ones, contain
disallowed words, or violate other outlined rules. Additionally, the system should provide a
probability score indicating the likelihood of a title being verified. Requirements: 1. Similarity
Check: a. Implement a mechanism to check for similar-sounding names using phonetic
similarity algorithms (e.g., Soundex, Metaphone). b. Identify titles that have common prefixes
or suffixes (e.g., The, India, Samachar, News). c. Ensure that variations in spelling or slight
modifications do not bypass the similarity check (e.g., Namaskar vs. Namascar). d. Calculate
a similarity percentage for each title comparison. 2. Prefix/Suffix Handling: a. Maintain a list
of disallowed prefixes and suffixes. b. Reject any new titles that include these disallowed
prefixes or suffixes if they cause the new title to resemble an existing title closely. 3.
Guideline Enforcement: a. Maintain a list of disallowed words (e.g., Police, Crime,
Corruption, CBI, CID, Army). b. Ensure that titles containing these disallowed words are
rejected. c. Prevent the creation of new titles by combining existing ones (e.g., if "Hindu" and
"Indian Express" exist, "Hindu Indian Express is not allowed"). d. Check for titles with similar
meanings in other languages and reject them (e.g., "Daily Evening" and "Pratidin Sandhya").
e. Disallow adding periodicity (e.g., daily, weekly, monthly) to existing titles to form new ones.
4. Verification Probability: a. Provide a probability score indicating the likelihood of a title
being verified. 5. Database Interaction: a. Efficiently search and compare new titles against
the database of 160,000 titles. b. Track current applications and use them for future
reference to reject similar titles submitted later. c. Use indexing and optimised search
techniques to handle the large dataset and ensure quick responses. 6. User Feedback: a.
Provide clear feedback to the user if their submitted title is too similar to an existing title,
contains disallowed prefixes/suffixes, violates guidelines, or is created by combining existing
titles. b. Display the verification probability to the user. c. Allow the user to modify their title
and resubmit it for verification. 7. Scalability: a. Design the system to handle an increasing
number of titles and user submissions. b. Ensure that the system remains performant as the
database grows. Expected Solution: 1. The system will provide the probability of a title being
verified. For instance, if a title has a similarity score of 80%, the verification probability shall
not be more than 100%-80%=20% 2. The system will reject any new title that is too similar to
existing ones, contains disallowed words or prefixes/suffixes, combines existing titles, or has
similar meanings in other languages. 3. The system will track current applications and use
them for future reference, rejecting similar titles submitted later by other users. Acceptance
Criteria: 1. Accuracy: a. The system correctly identifies similar-sounding titles and provides
consistent results. b. The system accurately rejects titles with disallowed prefixes, suffixes,
and words. c. The system prevents the creation of titles by combining existing titles and
identifies titles with similar meanings in other languages. d. The system disallows adding
periodicity to existing titles. e. The system provides an accurate verification probability score.
2. Performance: a. Title verification is completed within a reasonable time frame (e.g., under
2 seconds per title). b. The system can handle multiple title verification requests
simultaneously without significant performance degradation. 3 User Experience: a. Users
receive clear and actionable feedback on why their title was rejected. b. Users see a
probability score indicating the likelihood of their title being verified. c. The interface for title
submission and feedback is user-friendly and intuitive. 4 Robustness: a. The system handles
edge cases and variations in spelling effectively. b. The system is resilient to errors and
provides meaningful error messages when issues occur.

[Link].: 9

Problem Statement Title: Standardizing Odd School Structures to Improve Educational

Policy Implementation and Resource Allocation

Description 1: ODD School Structure refers to schools that do not conform to the standard
categories defined by the Ministry of Education under the UDISE+ portal. These standard
categories are designed to align with the Samagra Shiksha framework, which aims to ensure
uniformity and consistency in the educational structures across the country. Schools falling
outside these categories, due to variations in grade configurations, are classified as having
""odd structures."" However, 145,012 schools across the country fall into the ""odd"" school
structure category due to variations in grade configurations. These odd structures pose
challenges for the uniform implementation of educational policies and schemes, such as
Samagra Shiksha. This misalignment is particularly significant in states like Goa, Mizoram,
West Bengal, and Kerala, where the majority of schools do not conform to the standard
categories.
Description 2: The problem of odd school structures affects the equitable distribution of
resources, effective policy implementation, and the overall quality of education. Schools with
odd structures are often excluded from the streamlined processes and benefits that come
with standardization. The national average of schools with odd structures is 14.28%, but in
some states, the percentage is much higher, complicating the delivery of educational
initiatives. The objective is to standardize these odd school structures by aligning them with
the defined categories under Samagra Shiksha, thereby improving the uniformity and
effectiveness of educational policies across the country. Innovative Solutions: A. AI-Driven
School Structure Analysis Tool: Develop software that uses AI to analyze school data and
automatically classify schools into standard categories or identify them as odd structures.
This tool can also suggest optimal restructuring options to align schools with the standard
categories. B. Standardization Support Platform: Create a platform that provides guidelines,
best practices, and resources for schools to transition from odd to standard structures. The
platform could include modules on infrastructure planning, grade reconfiguration, and
community engagement strategies. C. Dynamic Resource Allocation System: Develop a
system that allows for flexible resource allocation based on school structures. This system
would help in distributing resources more equitably by taking into account the specific needs
of schools with odd structures during the transition period. D. Progress Monitoring
Dashboard: Implement a real-time dashboard for monitoring the progress of schools as they
transition from odd to standard structures. This dashboard would provide insights to
policymakers and administrators, allowing them to track changes, identify bottlenecks, and
intervene as necessary. E. Interactive Stakeholder Engagement Portal: Develop an online
portal to engage with school administrators, teachers, and the community. This portal could
include forums, feedback tools, and support resources to facilitate the transition to standard
structures, ensuring that the process is inclusive and addresses local concerns.

Domain: IOT (Internet of Things)

[Link].: 10

Problem Statement Title: Hardware Inventory Management in Police Department

Description: Effective hardware inventory management is crucial for the police department to
ensure that all technological assets such as computers, communication devices, servers,
and other equipment are available, functional, and up-to-date. Proper inventory management
supports the department’s operational efficiency, readiness, and ability to respond promptly
to various situations. However, the police department is currently facing several challenges
in managing its hardware inventory, leading to inefficiencies and operational risks. The lack
of a centralized system results in decentralized records, inconsistencies, and difficulty
tracking assets across locations. Without centralized tracking, monitoring the status,
location, and usage of assets becomes error-prone. Inaccurate inventory records due to
manual entry errors, outdated data, and infrequent audits lead to missing or unaccounted-for
hardware, causing financial losses and disruptions. Inefficient allocation further results in
resource misallocation, underutilization, and increased operational costs. Poor maintenance
and lifecycle management lead to delayed servicing, higher downtime, and performance
degradation due to obsolete equipment. Non-compliance with regulatory requirements and
security vulnerabilities arise from untracked or outdated devices. Inefficient management
also causes unnecessary purchases, budget overruns, and planning difficulties. These
issues impact operations by causing delays, increased downtime, resource wastage, and
potential security incidents. To address this, the department should implement a centralized
inventory management system with real-time tracking through barcode or RFID
technologies. Regular audits and timely updates should ensure data accuracy. Dynamic
allocation and utilization monitoring can optimize resource distribution. Scheduled
maintenance and lifecycle tracking will enhance performance and reliability. Compliance
tracking and security protocols will reduce regulatory and cyber risks. Cost analysis and
accurate inventory data will improve budgeting and procurement efficiency. By deploying a
robust, technology-driven inventory management system, the police department can
significantly improve operational efficiency, ensure resource availability, reduce waste, and
strengthen its overall ability to maintain public safety and order.

[Link].: 11

Problem Statement Title: Road Transport Network Telematics Develop a telematics

solution to enable efficient trucking operations for the long haul to connect the country
through route optimization, live tracking and monitoring, optimal capacity utilization analysis
and to enable appropriate response.

Description 1: The Department of Posts operates 76 national routes under All India Road
Transport Network (RTN). Commercial trucks of 5-14 tonnes capacity ply on these routes
carrying parcels and mail. There are also state or intra-circle routes operated through a
combination of commercial and departmental trucks for carrying parcels and mail.
Description 2: The RTN trucks, operating on various routes, touch intermediate points,
where parcels and mail are off-loaded and loaded. The capacity utilization of the trucks
fluctuates depending upon the volume of parcels/mail being carried by it [illustration: On the
Ahmedabad-Ajmer-Jaipur route, Ahmedabad loads the truck up to 80% capacity for Ajmer.
At Ajmer, 20 % of the volume is offloaded, and 30% loaded, bringing total loading to 90%
capacity (80-20+30 = 90). Similarly, at Jaipur, 40% volume is offloaded and 50 % loaded,
bringing total loading to 100% (90-40+50= 100)]. Real time data is required for optimal
utilization of truck capacity and adequate planning. In order to monetize spare capacity,
integration with 3PL partners for capacity sharing through API (as described in the previous
point) is also required. RTN trucks traverse a fixed route with specific touch points. The
trucks may be delayed due to various factors, such as vehicle breakdown, poor traffic
conditions, detours, detention at one or more touch points etc. GPS, geofenced locking
should be enabled for any detours or delays so that alerts may immediately be triggered for
verification and corrective action in time. Live tracking of trucks, with points of delay is
required for efficient and timely movement of parcels and mail in the value chain. There
should be a system for dispatch and schedule management as per live data automatically
populated and analyzed in real time. Any changes or deviations should automatically flow
into the system so that schedules, which are downstream, automatically get updated and
adjusted to the delay in real time. All data should be captured and populated automatically,
and relevant MIS reports generated automatically with visualization where relevant.
Expected Solution: As described above the expected solution is based of GPS leading to a
strong GIS for the network for it to send alerts and prompt necessary adjustments on the
interlinked parts of the supply chain for every run of the truck.
[Link].: 12

Problem Statement Title: AI-Powered Delivery Post Office Identification System: The wide,
evolving delivery network of the Post Office makes it difficult for customers to write the
correct pin code on postal items for delivery. The Post Office also merges pincodes together
to mechanize and streamline delivery in the emerging volume and mix of mail handled at
different stages, including the point of delivery. An intelligent solution powered by AI is
needed to meet the dynamic design of the supply chain for both customers and operators
within India Post.

Description 1: The Department of Post in India boasts the world's largest postal network,
with over 165,000 post offices. Through nearly 19,000 pin codes across the country the
postal network manages the delivery of postal items across its network. One common issue
is the mismatch between the specified Delivery Post Office name and the area PIN code on
mail items, leading to delays and customer [Link] delivery network design
keeps on evolving as Nodal Delivery Centres and Delivery Hubs are introduced to bring in
more efficiency in operations. From the perspective of the customer, the challenge is two
fold: (a) to know the correct Pin Code, which is itself a challeng; (b) the dynamic nature of
operations inside with merged pin codes etc for all or certain categories of mail is also a
challenge. Though the customer does not need to bother about (b), the internal system of
India Post needs to take care of both issues (a) and (b) for efficient and correct delivery at all
times.

Description 2: Postal services are essential for various government, financial, educational,
corporate, and public entities, both in urban and rural areas. Despite the widespread use of
postal services, inaccurate addressing persists, hampering efficient delivery. Each mail item
must clearly indicate the recipient's Delivery Post Office name and area Pin Code. However,
inconsistencies in this information often result in misrouted mail, with approximately 5% of
articles handled daily at distribution hubs containing incorrect Pin Codes. Even if correct
Pincodes are provided as per information available with the customer, there is a mismatch in
the internal operations as the constant upgradation of delivery network with Nodal Delivery
Centres with merged pincodes etc makes it too dynamic for internal operators to keep pace
with the changes. This leads to mistakes, delays and wasteful movement of mail pieces in
the system. Expected Solution: An AI-based scanning solution is proposed to identify the
correct delivery location for mail items with unclear or mismatched addresses. This system
will utilize advanced machine learning algorithms to analyze and interpret address
information, determining the appropriate Delivery Post Office based on available data. By
automating the identification process, this solution aims to improve last-mile delivery
efficiency, reducing delays and enhancing customer satisfaction. Having taken care of the
first problem, the system is also expected to align with internal operational databases for
clubbing and mapping of different pincodes for delivery purpose and thereby suggest the
right destination for dispatch and delivery of the articles.
Domain: Education & Learning

[Link].: 13

Problem Statement Title: AI-Driven Inclusive Assessment Tools for Skill Ecosystem

Description: An AI-powered quality assessment tool is proposed to evaluate candidates

comprehensively across India’s diverse skill ecosystem, encompassing pen-paper exams,
online MCQs and descriptive tests, practical exams, and viva voce assessments. This tool
will cater to all learners, including Persons with Disabilities (PWD), and support online,
offline, and blended modes of examination. In India’s vast skilling landscape, maintaining
fairness, inclusivity, and quality in assessments across multiple levels—from schools and
ITIs to specialized vocational training—is a major challenge. The AI-driven assessment
system aims to overcome these issues by offering a unified, intelligent, and adaptable
evaluation platform. It will provide holistic assessment coverage for a wide range of skills
and educational levels, ensuring comprehensive evaluation of cognitive, practical, and
communication competencies. Designed with inclusivity in mind, the system will feature
voice-to-text, text-to-speech, alternative input methods, and customizable formats to
accommodate PWD candidates effectively. Its multi-mode assessment capability will ensure
flexibility, enabling candidates from remote or underserved areas to participate seamlessly.
Using adaptive AI algorithms, the system will personalize tests in real time by adjusting
question difficulty and type according to the candidate’s performance. Advanced analytics
and feedback mechanisms will provide detailed insights into strengths and improvement
areas, helping educators and institutions optimize their training approaches. The tool will
also promote standardization and quality assurance, maintaining consistency and credibility
across regions and programs. Robust data privacy and security protocols will safeguard
candidate information and assessment results. The expected outcomes include enhanced
quality and fairness of assessments, equitable inclusion for PWD candidates, flexible and
scalable assessment solutions across different environments, actionable data-driven insights
for educators, and standardized evaluation practices throughout the skill
ecosystem—ultimately strengthening the quality, accessibility, and reliability of skill
development assessments in India.

[Link].: 14

Problem Statement Title: AI-Enhanced Career Guidance System for Personalized Career
Pathways

Description: Develop an AI-powered career guidance system that provides personalized

career pathways for students and professionals. The system should consider an individual’s
aptitude, aspirations, abilities, and work experience to recommend tailored career options
and future progression opportunities. Background: Career guidance is a critical component
of educational and professional development, yet many individuals struggle to find the right
career path that aligns with their skills, interests, and long-term goals. Traditional career
counseling methods often lack personalization and may not fully account for an individual’s
unique profile, including their aptitude, aspirations, abilities, and past experiences. This
hackathon challenge invites participants to create an AI-enhanced career guidance system
that leverages data to provide personalized, dynamic, and future-oriented career
recommendations for students and professionals.
Key Areas to Address:

Aptitude Assessment: The system should include AI-driven tools to assess an individual’s
natural aptitudes and strengths, identifying the areas where they are most likely to succeed
and find satisfaction in their work.

Aspirations and Interests: Incorporate methods to capture and analyze the user’s career
aspirations, interests, and values, ensuring that the recommendations align with their
long-term goals and passions.

Ability and Experience Mapping: The system should evaluate the user’s current abilities,
skills, and experiences, mapping these against potential career paths to identify where they
stand and what further development might be needed.

Future Progression and Skill Gaps: Use predictive analytics to identify potential future career
progression opportunities based on industry trends and individual growth potential. The
system should also highlight any skill gaps and suggest targeted learning opportunities to
help users advance.

User-Friendly Interface: Develop an intuitive, user-friendly interface that makes the career
guidance process accessible and engaging for users at all levels, from students exploring
initial career options to professionals considering a change or advancement.

Expected Outcomes

- Personalized Career Recommendations: The system will deliver highly personalized career
recommendations that take into account the user’s unique profile, including their aptitudes,
aspirations, abilities, and experiences.

- Enhanced Career Satisfaction: By aligning career recommendations with the user’s

strengths and interests, the system will help individuals find more fulfilling and satisfying
career paths.

- Improved Career Progression: The system will provide users with clear, actionable
guidance on how to progress in their chosen careers, including identifying skill gaps and
recommending targeted learning or development opportunities.

- Informed Decision-Making: Users will be empowered to make more informed career

decisions, with a clear understanding of the opportunities available to them and the steps
needed to achieve their goals.

- Scalable and Adaptable Solution: The AI-enhanced career guidance system will be
scalable across different educational levels and professional stages, making it a valuable
tool for a wide range of users, from students to mid-career professionals.
[Link].: 15

Problem Statement Title: Mentor Connect

Description:

Project Concept: Mentoring - The best way Mentoring during the Career/Education is a vital
aspect for the success of a candidate and this can achieve amazingly positive changes in
the life of a mentee. Today, India is known for its demographic advantage and career
directions are infinite. However, newer directions are demanding candidates to be coached
and mentored for progress. This application or platform can be developed by students
understanding this knowledge development need and bring together mentors and mentees.
Create a platform where candidates can connect with mentors based on the availability of
the mentors, and mentors should be able to help candidates with their queries and concerns
around careers, skill development, etc. Mentors could be senior leaders from the industry or
could be subject matter experts who have spent appreciable time in the industry and are
aware of intricacies around jobs in a specific industry. Requirements:

1. Create an automated calendar booking system: - Booking Automation: Develop an

automated calendar booking system that could create bookings automatically with a mentor
based on the mentor’s availability and schedules appointments with mentors based on their
availability. The system will check mentors' calendars and automatically find and book
suitable time slots for meetings. - User-Friendly Interface: Provide a simple, intuitive
interface for candidates to select their preferred time slots, ensuring easy and efficient
scheduling. - Integration with Calendly: Consider integrating with existing tools like Calendly
to leverage their robust scheduling capabilities.

2. Embedded video call feature: - Video Call Integration: Embed a video call feature within
the platform to facilitate virtual meetings between mentors and candidates. - Chat
Functionality: Incorporate a chat feature within the video call to allow text-based
communication during the session, enabling the sharing of links, documents, and other
resources. - Secure and Reliable: Ensure the video call and chat functionalities are secure,
reliable, and easy to use.

Expected Outcome: The platform will help connect candidates with mentors from the
industries and thereby increase their chances of employment through expert guidance and
referrals. NOTE: This is an innovation opportunity and students are encouraged to think out
of the box to develop solutions which can be presented in newer ways or which can address
the needs in out-of-the-box ways for a certain industry or make the platform generic. The
above description serves as a guide to specify essential needs that can be satisfied for the
developed solution and is not limited only to the scope described.

Domain: Sustainability & Smart City

[Link].: 16

Problem Statement Title: Bridging the Measurability Gap - A Digital Solution for validated
Citizens Charter norms' adherence across public interfaces and customer touchpoints of
DoP
Description 1: The Department of Posts (DoP) prioritizes citizen satisfaction with its
"Citizens' Charter" outlining service delivery standards. However, ensuring consistent service
quality across a vast network of operative offices remains a challenge. Real-time
measurability and transparent communication are crucial for building trust and accountability.
This innovative solution can transform DoP's Citizens' Charter monitoring system, leading to:
Increased Accountability and Transparency: Enhanced citizen trust through readily available
data on service delivery performance. Data-Driven Decision Making: Identifying areas for
improvement based on real-time insights. Improved Service Delivery: Proactive measures to
address delays and bottlenecks, leading to faster delivery times and higher citizen
satisfaction. By fostering a culture of data-driven decision making and citizen engagement,
this solution can solidify DoP's commitment to its Citizens' Charter and elevate its service
delivery to new heights.

Description 2: The challenge lies in developing a digital platform that streamlines and
enhances the measurability of Citizens' Charter compliance in DoP offices. This platform
should enable DoP to: Track Service Delivery Standards: Capture data on key performance
indicators (KPIs) defined in the Citizens' Charter, such as delivery timelines, customer wait
times, and complaint resolution rates. Standardize data collection procedures across all
operative offices for accurate and reliable information. Real-time Performance Monitoring:
Implement a real-time dashboard accessible to DoP officials and citizens. This dashboard
should provide a dynamic view of KPI performance across different regions and offices.
Utilize data visualization tools to present complex information in easy-to-understand formats
like charts and graphs. Enhanced Transparency and Communication: Allow citizens to track
the status of their postal services (e.g., mail delivery, money order processing) using a
unique tracking ID through the dashboard. Enable DoP to proactively communicate with
citizens regarding any potential delays or service disruptions through the dashboard and
other channels (SMS, email). Expected Solution: Seeking a mobile-first, cloud-based
platform with the following functionalities: Mobile App for Data Collection: Develop a
user-friendly mobile app for operative office staff to capture real-time data on Citizen's
Charter KPIs. Data Aggregation and Analysis Engine: Build an automated system to collect
data from the mobile app, integrate it with existing DoP systems, and analyze it for
performance insights. Interactive Dashboard Design: Create a visually appealing and
interactive dashboard accessible through web and mobile platforms for DoP officials and
citizens. Citizen Tracking Feature: Integrate a secure citizen portal within the dashboard for
tracking postal service progress using unique tracking IDs. Alerting and Notification System:
Design an automated system to send SMS and email alerts to citizens regarding potential
service disruptions or delays.

[Link].: 17

Problem Statement Title: Dashboard for Swachhta and LiFE. Develop a dashboard aimed
at maintaining cleanliness and LiFE practices, integrating AI-powered image processing
technology for effective monitoring of Swachhta and green practices adopted in post offices.
The dashboard needs to be accessible from the Divisional Office offering surveillance
capabilities and triggering alerts for deviation from prescribed Swachhta and Green Growth
standards to prompt on ground intervention to guide the post office concerned to make
necessary changes to conform to set expectations.
Description 1: The Department of Posts has implemented Swachhta Abhiyaan in its entire
network with intense and focused efforts in the last 10 years. In 2023-24 it has introduced
more systematic efforts towards institutionalization of Swachhta Protocols in its entire
network and also introduced Lifestyle for Environment Practices through an introductory
course on the iGOT portal of GOI and Mission Karmayogi portal of the Department. In the
year 2024, the Department has prescribed a Swachhta Action Plan (SAP) 2024 template
and provided a toolkit to the field units to adopt fit to local context activities under Swachhta
Abhiyaan from the menu of activities prescribed for the entire network to choose from.
Through the SAP 2024, the Department aims to institutionalize Swachhta protocols across
the network with local ownership and adherence to basic standards. To ensure the outcome
the solution to the Problem Statement may be useful for the Department to meet its
objectives across the network. In addition to the Swachhta Action Plan and related
behaviours and practices, the Department is also committed to adopt Lifestyle for
Environment Practices for each of its administrative and operative units. Monitoring system
for LiFE with modern and emerging technologies is also welcome to make the Swachhta
focused solution above even more useful and powerful for the Department.

Description 2: This is a problem statement seeking monitoring and data analytics solution
using modern and emerging Industry 4.0 technologies. While the use of AI and pictorial data
based automated monitoring is at the heart of the problem, the team working on this problem
can choose any combination of available technologies and tools to devise a simple to use
and adopt solution to enable its mass adoption across the post office network. Expected
Solution: The expected solution is pictorial data and analytics based on AI. Cameras to
capture pictorial data are available in various locations in the DoP and actual or dummy data
for the solution will be provided. The geolocation and timestamp on the pictorial data is also
to be taken as a fundamental data point to capture in the spirit of the outcome expected
towards improved internal governance of Swachhta and green, LiFE, practices being
adopted by the Department. The solution developed here could have a wide application
based on the scalability that is built in by the teams working on this problem statement.

[Link].: 18

Problem Statement Title: Platform for Inter-departmental cooperation (at city level) in
Indian Cities, for sharing data & resources, unified phasing, planning and implementation of
projects.

Description 1: Background: Multiplicity of authorities and implementing agencies in Indian

Urban landscape is one of the biggest challenges for urban governance. Not only does it
lead to underutilization of resources available with various departments, execution of
multiple projects at same site may also lead to interference and delays due to
miscoordination. A digital platform for interdepartmental cooperation can potentially resolve
these issues and make inter-departmental cooperation more streamlined. Example- A road
developed by one agency, may get damaged due to installation of gas pipelines or utility
ducts on road 2 months later. This can be resolved with preemptive collaboration between
both implementing agencies.
Description 2: The digital platform can provide solutions for-

1. Exchange of data and resources (technical expertise, machinery, technology etc.), and list
of ongoing and upcoming projects between various agencies.

2. Scheduling of tasks, reports, and work distribution between agencies for projects
categorized as "Inter-departmental"

3. Identification of projects in different departments which share the project site, and tools for
assistance in organizing meetings of various departments, unified project phasing (to reduce
costs incurred), planning and execution of projects.

4. Training and capacity building exercises, workshops & seminars.

5. Discussion forum with (Intra department, inter department, and public form sections.
Expected Solution- A digital platform for interdepartmental cooperation which allows
functions of- a. Registering departments, employees, officers, technical experts etc. with
their accounts. b. Sharing data, ongoing projects with location, and inventory of available
resources between departments/agencies. c. Creation of tasks, scheduling, reporting and
work distribution across departments for inter-departmental and multi-departmental projects.
d. Identification, unified phasing, planning, and implementation of projects on same site e.
Discussion forum, with different forum sections allowing access to specific groups of users.

Domain: Cybersecurity

[Link].: 19

Problem Statement Title: Improving open-source software security using Fuzzing

Description 1: Fuzzing is an automated process of identifying software vulnerabilities by

supplying unexpected and faulty inputs to the software. The main aim of fuzzing is to identify
the crucial edge cases where a software might fail. Therefore, fuzzing provides a crucial
insight into the stability and security of the software. The process of fuzzing can be divided
into following broad steps:

1. Identification of Target Function(s) Target function(s) are typically those functions that act
as entry points for processing input data. They use various APIs to perform operation on the
input data.

2. Developing harness Harnesses are small code stubs whose sole purpose is to invoke the
target function by using mutated data inputs. A harness bridges the gap between how the
fuzzer generates input and how the target application receives and processes the input.

3. Fuzzing - In this step, a fuzzer is used to generate numerous data inputs which are then
passed to the target function using the harness. The fuzzer checks whether the application
crashes by processing a certain input. If a crash occurs, then it saves the input and the
memory state of the crash to file for later analysis.

Description 2: Fuzzing has proven its effectiveness in discovering thousands of

vulnerabilities in file-processing and stateless applications. In fuzzing, and automated testing
in general, designing test oracles is crucial. In this challenge the team is supposed to fuzz an
open source software namely the Windows variant of Sumatra PDF Reader software
(version 3.5.2 or later). Sumatra PDF Reader is a very popular open source and widely used
PDF viewing software. In this challenge, teams are required to develop a working harness
for fuzzing of the latest version (version 3.5.2 or later) of Windows Sumatra PDF Reader
software solution, fuzzed on any fuzzer of their choice.

The submission will be evaluated on the following criteria: 1. Target functions identified 2.
Live demonstration of fuzzing harness developed 3. Code Coverage achieved 4. Technical
report submitted by the team.

Expected Solution: Each team must provide a fuzzing harness that is capable of fuzzing the
windows software solution of the Sumatra PDF Reader (version 3.5.2 or later). This fuzzing
harness must identify target functions and supply appropriate arguments for the invocation of
such functions. The fuzzing harness will be run using a fuzzer (preferably WinAFL). Each
team must submit a working harness along with a technical report stating: 1. Reversing
steps undertaken 2. Target functions identified 3. Dependencies identified

[Link].: 20

Problem Statement Title: Fake social media accounts detection

Description 1: At present the ITBP guards 3,488 km long India-China borders ranging from
the Karakoram Pass in Ladakh to Jachep La in Arunachal Pradesh. Apart from this, the
Force also has important roles in many internal security duties and operations against the
left wing extremism in the state of Chhattisgarh. Creating fake accounts on Facebook,
Instagram or at any other platform and fake account uses, should be identify for account
suspension or legal action. To safeguard the organization from the unknown fake account
messages over any social sites, a tool may be developed for their identification. Also a
central agency should be designated to get the information about the identified fake account
holder informed by the developed tools and accordingly, concerned social site organization
will approach to delete/suspend the fake account in time bound manner worldwide.

Description2: How to discover/identify fake profiles on Facebook, Instagram, twitter or other

social apps using tools. Also subsequently how to ignored/reported/make to delete of these
identified fake accounts by the tools/ through concerned agency in India.

Expected solution:

1. Tools for identifications of fake account should be developed for popular social sites.

2. A Central Agency should be designated to get the information about the identified fake
account holder info by the developed tools and accordingly, concerned social site
organization will approach to delete/suspend the fake accounts in time bound manner
worldwide.

[Link].: 21

Problem Statement Title: Money transacted by criminals in cryptocurrency

Description: In order to avoid using the banking system, the criminals also transfers the
defrauded money in crypto currency.

Challenges: Unable to find Beneficiary details of crypto transactions, and by whom this
amount was used, that information is not traceable.

Expected Solution: Create a mechanism to record peer to peer crypto transaction by crypto
exchanges.

[Link].: 22

Problem Statement Title: Domain hosting details

Description: On searching many suspicious websites on whois, their hosting appears to be

on cloud flare, but after sending mail to cloud flare, say that the said website uses one of
their services like SSL certificate etc. and they do not have the hosting of the website, due to
which correct information about the suspicious website cannot be obtained.

Challenges: Correct information from suspicious websites is not available. Expected

Solution: A mechanism should be developed that can provide accurate information about
suspicious websites along with their actual hosting.

Domain: Healthcare

[Link].: 23

Problem Statement Title: Ideas that can boost fitness activities and assist in keeping fit.

Description: In today’s fast-paced lifestyle, maintaining regular fitness activities and staying
physically active have become increasingly challenging for individuals of all ages. Sedentary
habits, lack of motivation, limited access to fitness resources, and poor time management
contribute to declining physical health and overall well-being. There is a growing need for
innovative solutions that can inspire people to adopt and sustain healthy routines through
technology, gamification, social engagement, and personalized fitness tracking. The goal is
to design creative and accessible approaches—such as AI-powered fitness recommendation
systems, virtual trainers, smart wearables, or community-based activity platforms—that
encourage daily exercise, monitor progress, and provide real-time feedback. Such solutions
should aim to make fitness engaging, inclusive, and adaptable to individual goals, ensuring
that users stay motivated and committed to improving their health and lifestyle over time.

Expected Solution: The proposed solution should leverage technology and behavioral
science to encourage consistent fitness engagement and promote long-term healthy habits.
An AI-based fitness companion app or platform can be developed that analyzes user
profiles, lifestyle patterns, and fitness goals to generate personalized workout and nutrition
plans. Integration with smart wearables or mobile sensors can track metrics like steps,
calories, heart rate, and activity duration, providing real-time feedback and motivation
through gamified challenges, rewards, and progress badges. The system can also
incorporate virtual coaching, social leaderboards, and community challenges to enhance
user accountability and enjoyment. Additionally, features like daily reminders, mental
wellness tips, and adaptive fitness programs can help users maintain a balanced approach
toward health and well-being. The ultimate objective is to create a comprehensive, engaging,
and intelligent fitness ecosystem that helps users stay active, track progress, and achieve
sustainable fitness outcomes.

[Link].: 24

Problem Statement Title: Software for Speech Language Therapy Clinical Services

Description 1: The rehabilitation institutes involved in this activity face high therapy cases
load and more number of student therapists with less number of therapy supervisors.
Following this system in manual mode impedes providing prompt services.

Description 2: Clinical services with respect to speech language therapy begins with patient
allocation to the therapist. The therapist then prepares therapy plan with goals and activities.
The respective case supervisor needs to check the therapy plan and provide the inputs for
further continuation of regular therapy sessions. At least post ten therapy sessions the
therapist writes the progress report of the therapy case. Further, it is evaluated by the
supervisor. This Cycle of speech language therapy clinical services in continuous until the
therapy case has completely improved or discontinued due to various reasons. The
supervisor based on overall handling of therapy sessions gives clinical ratings.

Expected Solution: The process needs to become digitized with the use of software where in
right from the allocation of cases, therapy related documentation by both therapist and
supervisor, clinical evaluation of therapy, feedback and clinical rating can be performed.

[Link].: 25

Problem Statement Title: Queuing models in OPDs/ availability of beds/ admission of

patients. A hospital based solution is ideal which can be integrated with city wide module

Description: Technological solution as per queuing models in OPDs/availability of

beds/admission of patients would be one area. Study of dispensation of various types of
medicines/consumables and Inventory management modules at hospital level are key areas
requiring support. NIC has already developed some modules but their implementation in
Delhi is yet to be started. A hospital based solution is ideal which can be integrated with city
wide module may be required.

Domain: Open Innovation

[Link].: 26

Problem Statement Title: Develop software that converts 2D blueprints of buildings into 3D
walkthrough models for NSG operations, enhancing situational awareness and efficiency.

Description:

1. National Security Guard (NSG) operates inside buildings, built up areas and other
infrastructures like metro stations, schools, etc in case of any terror. Since the operations are
of National security significance, better understanding of the target layout is of paramount
importance. 3-Dimensional walk through of the buildings helps in the efficient understanding
of the target layout and also enables effective briefing of the troops by the Commander for
operating inside such targets.

2. Presently, the building authorities provide 2-Dimensional layout of the building in the form
of blue prints/ floor plan. Assimilation of the situation and appreciating the potent threat with
the help of 2-Dimensional layout turns out to be a challenging task. In such a scenario, a
3-Dimensional walk-through of the building helps significantly in increasing the efficiency of
the ground forces while operating.

3. The software which needs to be developed should take the 2-Dimensional blueprint layout
as input and also should take various parameters like length, breadth, height of the building,
layout of staircase, entry and exit, etc as additional inputs. The software should then export
them to a file and create a full fledged 3-Dimensional walkthrough model of the building. The
software must also include the facility for integration of exported model to the offline Google
maps, Satellite pictures (if any available).

4. Since the target includes buildings which are secretive in nature, the software is required
to be operable fully in standalone mode, without any internet connectivity required in future.
For the initial installation phase, the system can be connected to the internet to facilitate
fetching up of Maps, satellite imagery from open source, etc required for the software
functioning.

5. This software would also reduce the time taken during the initial briefing of the troops prior
to the launch of any operation. It further increases the efficiency of the operations, thus
resulting in early success.

6. The conversion of 2-Dimensional blueprints to 3-Dimensional models must address

following problems. (a) Software should take 2-Dimensional blueprint as an input and
produce 3- Dimensional walk through as output. (b) Provisions to input parameters like
length, breadth, height of the building and other details should be included. (c) Minimum
supervision from the user end should be needed to finalise the conversion. (d) Offline
integration with Google maps, satellite imagery to get enhanced output covering the nearby
geographical areas as well.

[Link].: 27

Problem Statement Title: Development of a Geolocation-Based Attendance Tracking

Mobile Application.

Description: A mobile application needs to be developed to streamline and automate the

attendance tracking of its employees across multiple office locations. This application will
leverage geolocation technology to record employee check-in and check-out times along
with geo-location as they enter and leave the office premises. Aim of this App is to enhance
operational efficiency, reduce manual attendance tracking errors, and provide a seamless
experience for employees to manage their work-related movements.

Key Requirements:
1. Geolocation-Based Check-In and Check-Out: Automatically record check-in time and
geo-location when an employee enters within a 200-meter radius of his office. Record
check-out time when the employee leaves the 200-meter radius. Ensure that each check-in
is paired with a corresponding check-out, regardless of the frequency of entries and exits.

2. Manual Location Check-In / Check-Out for Offsite Work: Provide a manual check-in and
check-out feature for employees working at locations other than his offices. Automatically
suggest relevant locations based on real-time longitude and latitude data, allowing
employees to confirm their check-in and check-out at these suggested locations.

[Link] the employees total working hours.

4. Data Accuracy and Integrity: Ensure the application maintains accurate and tamper-proof
records of all check-in and check-out events. Enable real-time data synchronization and
secure storage to prevent data loss and ensure reliability. The teams should use free and/or
open-source resources, as applicable, for the entire project.

[Link].: 28

Problem Statement Title: Women Safety Analytics - Protecting Women from safety threats

Description 1: The growing concern for the safety of women and the increase in crimes
against women in various cities, highlight the need for advanced surveillance and analytical
solutions to protect women from various possible threats. We need a promising approach to
address these issues through real-time threat detection software.

Description 2: By leveraging advanced analytics through real-time monitoring, Women

Safety Analytics should create safer environments for women and assist law enforcement in
effectively addressing and preventing crimes against women. The proactive approach of
detecting anomalies and generating alerts can play a crucial role in enhancing public safety
and fostering a secure atmosphere for women. Women safety analytics software should
continuously monitor the scene to count the number of men and women present, offering
insights into gender distribution in specific locations and times. It should identify unusual
patterns, such as a lone woman at night, unusual gestures and generates alerts to pre-empt
potential incidents. Advantages of the system: ? By providing real-time monitoring and alerts,
the system helps to create a safer environment for women. ? Early detection enables law
enforcement to intervene before situations escalate. ? Continuous analysis provides
valuable data to identify hotspots and trends, aiding in strategic planning for city safety

Expected Solution: Women safety analytics should include the following functionalities

1. Person detection along with Gender Classification

2. Gender Distribution : Count the number of men and women present in the scene

3. Identifying a Lone Woman at Night time

4. Detection of a Woman Surrounded by Men 5. Recognizing SOS situation through gesture
analytics

6. Identifying hotspots where incidents are more likely to occur, based on the past alerts