Design of system for parking and repositioning of bikes in a public bike-sharing system

• Main Author: Siong, Jun Wei
• Other Authors: Sivakumar Appa Iyer,Siva
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2020
• Subjects: Engineering::Mechanical engineering
• Online Access: https://hdl.handle.net/10356/141046
• Institution: Nanyang Technological University

This FYP Report evaluates the current bike-sharing systems – the bike-sharing system with docking stations and dockless bike- sharing system – to understand the advantages and disadvantages of each system. The author then proposed a new bike-sharing system to address the shortcomings of the two current systems while retaining the advantages of them. The new bike-sharing system includes a proposed chain lock which offers a hybrid between the locking mechanism currently present in the bike-sharing system with docking points and dockless bike-sharing system. This proposed lock brings about a myriad of advantages such as being space efficient, has a low cost of infrastructure, self-generating docking points, promotes a more orderly parking of bikes, deter bike theft and has a lower cost of production. The author also recognised that profitability is the key to a sustainable business. Thus, it is important for bike-sharing operators to optimise their profits. As such, this project proposes a new mathematical model for profit optimisation. The proposed model makes uses of the results of the Newsvendor model, the concept of the Wagner-Whitin model and accounts for returning bikes to a bike station. The Newsvendor model determines the optimal quantity of bikes to be placed at each time period in order to maximise profits. Next, the concept of the Wagner-Whitin model is used to determine the bike reallocation plan that incurs the least cost. Lastly, the proposed model accounts for the returning bikes, which is important as it will affect the inventory and thus, inventory cost, from one period to another. In the case study of this report, the proposed model is shown to be able to determine a bike reallocation plan that incurs the least cost. Also, the model provides information of the quantity of bikes to reallocate and when to reallocate to incur the least cost. Furthermore, the analysis of the model is carried out on an hourly basis, which is critical as the demand for bikes are highly time-dependent, thus, allowing a more representative analysis.