| Project Id
|
BRJP26100031
|
| Project Detail
|
| Project Title
|
AI-Driven Urban Safety and Emergency Response Optimization Using Video Analytics and Predictive Data Science
|
| Senior Supervision Team (BITS)
|
| Supervisor name and Title
|
Prof. Maripini Himabindu
|
School or Department (or company, if applicable)
|
BITS PILANI, PILANI CAMPUS
|
|
|
|
Email ID
|
maripini.himabindu@pilani.bits-pilani.ac.in
|
| URL for more info
|
https://www.bits-pilani.ac.in/pilani/maripini-himabindu/
|
| a) Are you currently supervising a BITS or RMIT HDR student?
|
NO
|
| Please comment how many you are supervising
|
|
| b) Have you supervised an offshore candidate before?
|
NO
|
| If no, what support structures do you have in place?
|
|
| If yes, please elaborate
|
|
| Senior Supervision Team (RMIT)
|
| Supervisor name and Title
|
Nirajan Shiwakoti
|
School or Department (or company, if applicable)
|
STEM
|
|
|
|
Email ID
|
nirajan.shiwakoti@rmit.edu.au
|
| URL for more info
|
https://www.rmit.edu.au/profiles/s/nirajan-shiwakoti
|
| a) Are you currently supervising a BITS or RMIT HDR student?
|
NO
|
| Please comment how many you are supervising
|
|
| b) Have you supervised an offshore candidate before?
|
NO
|
| If no, what support structures do you have in place?
|
|
| If yes, please elaborate
|
|
| Other Supervisors (BITS)
|
| Supervisor name and Title
|
BANDHAN BANDHU MAJUMDAR
|
School or Department (or company, if applicable)
|
BITS PILANI, HYDERABAD CAMPUS
|
| Phone Number (Optional)
|
+919647793654
|
Email ID
|
majumdar@hyderabad.bits-pilani.ac.in
|
| URL for more info
|
https://www.bits-pilani.ac.in/hyderabad/dr-bandhan-bandhu-majumdar/
|
| Other Supervisors (BITS)
|
| Supervisor name and Title
|
Seyed Mojib Zahraee
|
School or Department (or company, if applicable)
|
STEM
|
| Phone Number (Optional)
|
+61421169347
|
Email ID
|
seyedmojib.zahraee@rmit.edu.au
|
| URL for more info
|
https://www.rmit.edu.au/profiles/z/seyedmojib-zahraee
|
| Field of Research (For Codes)
|
| 400512 | Transport engineering | 50.00 |
| 460207 | Modelling and Simulation | 25.00 |
| 460304 | Computer Vision | 25.00 |
|
| Project Description
|
|
Urban transport networks are essential to sustainable and liveable cities, yet urban intersections remain major safety and efficiency bottlenecks. Current safety management systems are largely reactive, relying on historical crash data that overlook numerous unsafe interactions and near-miss events that do not result in reported crashes. This PhD project proposes an AI-driven urban safety and emergency response framework that integrates video analytics, data science, machine learning, and intelligent traffic control to proactively reduce crash risk and improve post-crash response.
The study first applies advanced video analytics to extract detailed vehicle–vehicle and vehicle–pedestrian interaction data at intersections. Using AI-based detection and tracking, surrogate safety measures such as time-to-collision, post-encroachment time, deceleration rates, and near-miss frequency are quantified to reveal latent safety risks not captured in traditional crash databases. These metrics are spatially and temporally aggregated to generate intersection-level risk profiles.
In parallel, historical and real-time crash data are analysed to identify spatial and temporal crash patterns across the urban network. A dynamic dashboard is developed to visualise crash frequency, severity, and emerging hotspots, enabling near real-time monitoring of intersection safety. Machine learning models then predict crash-prone intersections by fusing aggregated risk profiles with explanatory variables including intersection geometry, control type, traffic volumes, pedestrian activity, lighting, weather, and temporal factors. Model interpretability methods are used to identify key contributors to predicted risk.
The dashboard serves as a decision-support tool by recommending targeted, factor-specific countermeasures such as geometric redesigns, signal timing adjustments, lighting improvements, enforcement strategies, and pedestrian safety interventions. Beyond prevention, the framework extends to post-crash response optimisation by informing the strategic placement of emergency response units. Following an incident, an AI-enabled adaptive signal pre-emption system dynamically coordinates signalised corridors to reduce emergency response and patient transport times while minimising disruption to general traffic. Overall, the project delivers an integrated, AI-powered urban safety framework linking predictive analytics, proactive interventions, and intelligent traffic control.
|
| Project Deliverable/Outcomes
|
|
Expected Outcomes
1. AI-driven reliable prediction of crash-prone intersections and hotspots using ML models trained on historical crash data and real-time traffic features.
2. Explainable identification of critical contributing factors for each hotspot, enabling site-specific interventions.
3. Near-real-time monitoring dashboard for traffic operators, visualizing crashes, near-misses, and high-risk intersections for data-driven decision-making.
4. Reduced emergency response time through predictive resource placement.
5. Automated, low-disruption emergency signal pre-emption
6. Network-wide improvements in safety, mobility, and resilience
7. Direct contribution to Sustainable Development Goals (SDG 3, 9, 11)
The work undertaken across the project stages will be documented and compiled into PhD thesis and subsequently refined for publication in high-impact journals. Year-wise Deliverables: Year 1 will focus on a comprehensive literature review, ethics approvals, data acquisition, and development of the video-based safety analytics and surrogate safety indicators. Year 2 will involve machine learning model development, crash hotspot prediction, model interpretability, and the design of the real-time safety analytics dashboard. The final year will deliver the emergency service deployment optimisation framework, adaptive signal pre-emption algorithms, full system integration and validation using real-world case studies, followed by final reporting, journal publications, and PhD thesis submission.
|
| Research Impact Themes
|
| SUSTAINABLE DEVELOPMENT AND ENVIRONMENT
| CLEAN ENERGY AND SUSTAINABLE TECHNOLOGIES |
| AI/ML and Data Analytics / Data Science with a focus on applications/translation in | Transportation & Logistics |
| ENHANCED LIVABILITY AND URBAN FUTURES | URBAN ENVIRONMENTS AND SMART CITIES, WATER STEWARDSHIP AND EFFECTIVE WATER USE |
|
| Which RMIT Sustainable Development Goal (SDG) does your project align to
|
|
SUSTAINABLE CITIES AND COMMUNITIES
|
| Which RMIT Enabling Impact Platform (EIP) does your project align to
|
|
URBAN FUTURES
|
| Which RMIT Program code will this project sit under?
|
|
DR216P23 PhD (Mech,Manu & Mech Eng)
|
| Student Capabilities and Qualifications
|
|
1. Strong interest and working experience in one or more of the following: urban mobility, safety, emergency response and smart city applications; 2. Proficiency in programming: Python, Machine Learning, Computer vision and video analytics, Data analytics
|
|
1. Experience with predictive modeling, spatio-temporal data analysis, or reinforcement learning for decision-making, 2. Experience with interdisciplinary projects involving multiple data sources and dashboard development
|
|
MTech
|
| Preferred discipline of Student
|
| Civil Engineering, Structural Engineering |
| Computer Vision, Image Processing, Virtual Reality |
| Data Science, Data Mining, Data Security & Data Engineering |
| Design, Design Engineering, Sustainable Design |
|