Research and Projects
There are 4 verticals in FedEx SMART Centre. Each one has it’s core research group having faculty members from IIT Madras.
- Supply Chain Sustainability
- Logistics Worker Wellness
- Algorithms and ML
- Logistics Infrastructure
P1
Digital and sustainable Supply chain Modelling and Analytics
Faculty members
Prof. Arshinder Kaur profile link
Prof. Chandrasekharan Rajendran profile link
Prof. R P Sundararaj profile link
Prof. Usha Mohan profile link
Prof. Vaibhav Chawla profile link
Prof. Nargis Pervin profile link
Focus area
Sustainability- Environmental and Net Zero
Objective
To explore how the logistics sector incorporates environmental and green practices toward achieving environmental, sustainable and efficient logistics.
Social Sustainability
Objective
To develop frameworks of socially responsible supply chains considering ESG factors, mental health and wellbeing
To understand how digital tools like blockchain and mobile applications can empower workers in supply chains, the research aims to assess their impact on job satisfaction, worker rights, and overall social well-being
Digital Transformation and Automation, Prediction and Optimization & Digital interventions
Objectives
To develop predictive models for lead time estimation and customer demand forecasting and capacity planning
To integrate planning, production, inventory, distribution and routing with end-to-end optimization solutions using AI and ML
To explore and develop prototypical systems using digital twins and digital platforms that can enable the seamless tracking, monitoring, and optimization of shipments, warehousing and packaging containers in global supply chains
Multi-modal and Global Logistics
Objectives
To assess the current state of global supply chains across different transportation modes, identifying inefficiencies and sustainability gaps.
To enhance collaboration among stakeholders including governments, industries, and international organizations to facilitate the implementation of efficient and sustainable global supply chain practices.
To enhance supply chain resilience by diversifying transportation routes and developing risk management frameworks
P2
Holistic Human machine collaboration with biomedical-cognitive measures in logistics and supply chain services
Vertical : Logistics Worker Wellness
Faculty members
Prof. Babji Srinivasan profile link
Prof. Rajagopalan Srinivasan profile link
Focus area
Objectives
Worker Health Management:
Monitor signs of physical fatigue related effects using bio-signal measurements
Estimate the cognitive workload, Vigilance level & mental fatigue using eye gaze parameters.
Worker Training:
Synchronize data from the process, automation system and Human activities
Notify with visual/audio alarms and assistance to rectify the shortcomings.
Create new policies/ training methods to prepare the workers better for the task
Human-Machine Teaming:
Develop algorithms for COBOTS that can resolve human-human and human-machine conflicts of actions.
Switch task/work between humans as well as between humans and machines based on situation awareness of humans and the significance of task/work.
P3
Accelerating Learning and Algorithms for Logistics Problems
Vertical : Algorithms and ML
Faculty members
Prof. Chandrashekar Lakshminarayanan profile link
Prof. Rupesh Nasre profile link
Prof. N.S. Narayanaswamy profile link
Prof. Anil Prabhakhar profile link
Prof. Rahul Marathe profile link
Focus area
The objective of this project is to develop algorithms, machine learning techniques and software modules that can be used by stakeholders in the logistics domain in order to improve overall efficiency of logistics operations.
Software system, Algorithms and API access to software implementations.
[Prof. N.S. Narayanaswamy]
Data and Analysis
Data generation
Simulation
Impact analysis
Tracking using Fastag data
Tracking using crowdsourced data
Consignment flow analysis
Customer feedback mining
Optimizations
Route planning and optimizations
Vehicle container space sharing
Container space packing
On-demand Asset sharing
Custom clearance optimization
Optimal Service points location finding
Rail network efficiency improvement
Cargo movement scheduling
Multimodal logistics operation & Transshipment
Platform
Integrated platform
Open Network and Framework for Logistics
Methods to implement KYC collection
Easy onboarding and verification
Prediction based
Fare validation
Demand forecasting
Employee requirement forecasting
Specific functionality
AI based logistics specific customer service
AI based eSRG
Skill training / upskilling through App
Seamless Communication System
Scalable solutions to Logistics problems using parallelization
[Prof. Rupesh Nasre]
CVRP (Capacitated Vehicle Routing Problem) is a combinatorial optimization problem, which is NP-hard and is of practical importance in the logistics industry. Plan is to utilize parallelism to improve the scalability of the CVRP computation
RL (Reinforcement Learning) for logistics – for example Railway Dispatching
[Prof. Chandrashekar Lakshminarayanan]
Because of the inefficiencies like low speed and preference given to passenger trains as compared to freight trains, the potential of railway logistics is not fully utilized in India. The goal here is to develop an RL based system that can provide recommendations for better efficiency of the rail network.
Quantum Computing based implementations to help us find better solutions to some of the NP-hard problems.
[Prof. Anil Prabhakhar]
Quantum Machine Learning will augment the software system through the use of hybrid quantum-classical algorithms that will enhance and hasten the search for feasible solutions to the NP-hard problems in route planning and optimization, and bin-packing.
P4
Development of Modules capable of integrating with current infrastructure to enable Autonomous Delivery Agents to adopt Advanced Delivery Vehicles
Vertical : Logistics Infrastructure
Faculty members
Prof. Satya R Chakravarthy profile link
Prof. Gitakrishnan Ramadurai profile link
Focus area
Adopting autonomous delivery vehicles like drones and rovers can greatly improve logistics management. However, current systems are fragmented among operators, logistics companies, and regulators, creating integration and coordination challenges. As the use of these vehicles grows, we need a unified system to ensure they comply with regulations and work well within existing frameworks.
Objectives
Building Autonomous Delivery Modules
Developing specialized modules for autonomous delivery agents within logistics operations.
Addressing Infrastructure Gaps
Focusing on bridging the gap in infrastructure between operators, logistics companies, and regulators to streamline operations.
Ensuring Regulatory Compliance
Prioritizing regulatory compliance and operational safety standards to meet industry requirements.
Enhancing Efficiency and Sustainability
Aiming to improve operational efficiency and reduce carbon footprint through innovative solutions.