Researcher

Mrs Sumana Biswas

Keywords

Fields of Research (FoR)

Autonomous agents and multiagent systems, Machine learning, Artificial intelligence, Decision making, Optimisation

Biography

Dr. Sumana Biswas is a confident, passionate devotee of Engineering with significant academic background. She is currently working as a Research Associate in Capability Systems Centre, School of Engineering and Information Technology at the University of New South Wales (UNSW), Canberra @ Australian Defence Force Academy, Australia. Dr. Biswas obtained her Ph.D. degree in Mechanical Engineering from the University of New South Wales (UNSW),...view more

Dr. Sumana Biswas is a confident, passionate devotee of Engineering with significant academic background. She is currently working as a Research Associate in Capability Systems Centre, School of Engineering and Information Technology at the University of New South Wales (UNSW), Canberra @ Australian Defence Force Academy, Australia. Dr. Biswas obtained her Ph.D. degree in Mechanical Engineering from the University of New South Wales (UNSW), Canberra, Australia. She received her B.S.C. and M.S.C. degree in Mechanical Engineering, from Chittagong University of Engineering and Technology (CUET), Bangladesh in 2006 and 2014 respectively.  

She has more than ten years of teaching and more than thirteen years of research experience. She was an Assistant Professor at the Chittagong University of Engineering and Technology (CUET), Bangladesh. She has supervised a number of undergrad and master’s students over these years. 

Her research interests include mission planning, path planning, collision avoidance, task scheduling, mission forecasting for autonomous agents and Artificial Neural Networks. In addition, her research focus is to solve technology decision-making problems in the field of capability management. She has published more than 25 peer-reviewed journals and conference papers. 


My Awards

1. Associate Fellow of Higher Education Academy (AFHEA) In recognition of attainment against the UK Professional Standards Framework for teaching and learning support in higher
education. Fellowship reference PR22387.

2. Dean's Award For Outstanding PhD Theses: Real-Time Path Planning for a Swarm of Autonomous Systems.


My Research Activities

1. Capability Model Development In this project I am working on the simulation-based optimisation model development project for an emergency military medical evacuation system. I have worked on a literature review paper on this topic. I have conducted an analysis of the literature review to investigate the decision problems involved in the total military medical evacuation process. I have also worked on identifying the emerging challenges of future MEDEVAC and the adoption of emerging technologies, emerging concepts, and advanced decision analysis methods in military medical evacuation to tackle these challenges. 

2. Consideration of Uncertainty in Dynamic Modelling System In this project, I have characterized the uncertainty for product family evolution. I have addressed the data uncertainty and parameter uncertainty of a dynamic modelling system using a state-of-the-art MCDM technique; Interval Valued Fermatean Fuzzy Multi-attribute Border Approximation Area Comparison (IVFFMABAC) method and forecasting through Deep learning. In this case, to predict the feature characteristic of future products, I have used different forecasting models such as LSTM, ARIMA and SVM.

3. Dynamic Modelling for Product Family Evolution In this project, I have proposed a Multi-Criteria Decision Making (MCDM) and Artificial Neural Network-based novel dynamic model for Product Family Evolution. I have proposed a new variant generation mechanism and developed an evolution graph for the product family. A product evolution graph is a diagram that shows the inheritance of genetic conditions among multiple generations. It is very significant to understand the evolution pattern from generation to generation. My proposed model incorporates the forecasting model with the evolution model which can help managers to become involved in decision making regarding the design of the future product.

4. Optimization to Solve RCPSP In this project, I have worked on optimization models for solving Resource-Constrained Project Scheduling Problems (RCPSP). In this research work, I have proposed a multi-objective particle swarm optimisation (MOPSO) based algorithm for solving project scheduling problems using MATLAB and PSPLIB library for RCPSP.

5. Multi-Agent Systems For a successful mission, efficient path planning with multi-agents is crucial. In this project, I have implemented a framework to investigate the dynamic path planning of a multi-agent autonomous swarm to execute the task of traversing a certain terrain with a specified start and goal state.

6. Coverage Planning For a candidate algorithm, it is difficult to handle multiple autonomous agents in dynamic environments, where it is necessary to replan a new path to perform the given tasks without any collision. To address this problem, a path-planning algorithm based on Particle Swarm Optimization is developed in this project to find an optimal path for the multi-agent system working under certain prescribed tasks.

7. Task Scheduling Task scheduling has become more and more vital in the field of mission planning since a wide variety of tasks need to be performed efficiently. In this research work, I have developed an effective solution to the task assignment problem combined with path planning using a group of autonomous agents.

8. Multiobjective Optimization and Forecasting In mission planning, before starting any mission, forecasting the optimal value such as the optimal number of vehicles or optimal distance required to accomplish a mission that takes into account the current circumstances, has become essential in one of today\textsc{\char13}s intelligent decision-making plan. I have proposed a Multi-Objective Optimization (MOO) approach.  To determine the required resources, the MOO approach can optimize both distances travelled and the number of vehicles for the mission planner. In addition, I have developed a data-driven neural network-based prediction model that will forecast the mission completion time with a reasonable accuracy which will utilize the historical information of the previous missions.


My Research Supervision


Areas of supervision

Autonomous system path planning, mission planning, Optimization, Simulation-Optimization, Technology decision making, Machine learning. 


Currently supervising

Interested to supervise undergraduate students and higher degree research candidates.


My Teaching

Decision Making in Analytics - ZZCA6510

Study level: Postgraduate

View less

Location

Canberra, ACT, Australia

Contact

+61 2 5114 5629