Research & Projects
Project 1
Free-floating bike repositioning problem with faulty bikes
As the docked bike-sharing system (DBS) differs widely from FFBS and in docked bike-sharing, bikes are parked only at the docking stations. During the repositioning operation, both types of bikes, usable and faulty, are supposed to be at fixed capacity docking stations.
Objectives
The use of FFBS has increased swiftly worldwide and many people are benefiting from it. However, there is always a deficiency or excess of sharing bikes in the system due to uneven flows that are either unable to accommodate users at some points or remain sedentary at some other points. Therefore, periodical repositioning of bikes in the system is inevitable to fulfill the commuters' demand at all service locations.
Results
The main contribution of the study conducted by me lies to present a framework for static FFBS repositioning with faulty bikes, where the framed model help bike-sharing companies: i) to perform usable bikes repositioning at different nodes, ii) in gathering faulty bikes from different locations to the depots, and iii) in the distribution of revamped bikes into the system.
Project 2
Towards An Energy Efficient Solution For Bike-Sharing Rebalancing Problems: A Battery Electric Vehicle Scenario
My second project was to formulate and design a model that focuses on the use of BEVs instead of using conventional ICEVs deployed for the rebalancing an FFBS.
Objectives
In today’s world, the health of economic growth is closely linked to the transport system. However, internal combustion engine vehicles (ICEVs) mainly power the current transport systems. This not only makes the world dependent on the global oil market but also generates the most important source of greenhouse gas (GHG) emissions. Due to the expected shortage of oil and increased emission of toxic gases, more and more talent and resources are being built up to meet the challenges of reducing oil dependence, climate change, and sustainable transport systems.
Results
My developed model formulation solves the problem in various situations. The results show that the vehicle battery capacity has a significant effect on the routing plan. A vehicle with small battery capacity visits the charging station after a short trip. The extra visits to the charging station not only increase operation cost but also increase the operation time. In addition, excessive battery charging events further delay the operation process. When the battery capacity is increased from 16 kWh to 60 kWh, the total traveling cost reduces up to 17.8% and 20.4% with one and two operating vehicles, respectively. Likewise, the cost of a vehicle with a capacity of 30 bikes is 44% lesser than the cost of the vehicle with a capacity of 15 bikes. Since the BEVs have a high initial cost. Therefore, a detailed cost analysis is required for various capacity vehicles.
Project 3
Physics-Informed Neural Networks (PINNS)-Based Traffic State Estimation: An Application To Traffic Network
Traffic state estimation (TSE) refers to the prediction of traffic state variables such as flow, density, and speed of road segments using partially observed traffic data.
Objectives
The traffic conditions on road segments in a network are usually described by macroscopic traffic state variables, such as flow rate, vehicle speed, and vehicle density as traffic streams. Transport planners identify congestion levels and traffic demands, as well as bottlenecks on roadways, through these indicators. However, these important measurements are not available for all locations or times due to physical and budget constraints of the measurements; even if the availability issue is not a concern; undesired noises in the measurements have been making ITS traffic management operations difficult. Combining factors such as the cost of sensor installation, the accuracy of vehicle detection techniques, and restrictions on data storage and transmission can often lead to partial observations of traffic state variables. To deliver effective traffic management, it is vital to estimate traffic state variables at locations without sensor data.
Results
The experimental results show that the presented framework accurately learns the complex traffic dynamics over a circular traffic network. The predicted speeds match the actual speeds at almost all the locations other than the intersection points. The error at the intersection points is due to discrepancies in the actual data. Theoretically, the flow conservation condition must be satisfied at the intersection points. However, the actual simulated data violate the flow conservation, and this violation is potentially the result of considering average speeds over small cells.