Scheduling and route planning of unmanned aerial vehicle for multi-parcel postal delivery in Singapore

• Main Author: Ee, Germaine Jia Min
• 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/10356/141261
• Institution: Nanyang Technological University

With rising popularity in online shopping, volume of parcels in Singapore has been increasing at an average of 14% every year and this trend is expected to continue for the upcoming years. In 2017 and 2018, Singpost has been fined for failure to meet delivery demand as it struggles to cope with the rising demand for home delivery. An alternative solution is needed to increase the manpower to cope with the current and future demand. Hence, UAV was introduced to increase the manpower capacity, especially in delivering lightweight parcels. As there are limitations on the total weight carried by UAV and the flight time of a UAV, it is important to maximise these constraints for efficient delivery using UAV. Hence, multi-parcel delivery by UAV was introduced. In a multi-parcel delivery by UAV, packing of parcels and planning of route are required to achieve optimisation in the above constraints. Yet, many research on multi- parcel delivery only focus on route planning to find the shortest route. Furthermore, many studies on scheduling only considered optimisation with weight limitation. Therefore, this project aims to develop a delivery system that focus on both scheduling and route planning, where there is a limitation on the number of items and total weight each UAV can carry in a trip during scheduling. The process flow of the UAV delivery system were discussed. Mathematical model and algorithm on scheduling and routing planning were developed. Power consumption of UAV was also modelled as a linear regression line. Results showed that delivery by UAV was optimised. There was also a minimal difference of 2% between the linear regression line and the power equation. The delivery system has also achieved a significant reduction in time and cost with at least 20% improvement as compared to the current delivery system. Therefore, a feasible multi-parcel UAV delivery system in Singapore was achieved.