Mathematical models and approximation procedures for vehicle routing problem with safe manual materials handling

Name: Prachya Boonprasurt

keyword: Vehicle routing problem (VRP)

LCSH: Vehicle routing problem -- Mathematical models

; Workload

; Manual material handling

; Genetic algorithm (GA)

; Hybrid GA

Abstract: Since most vehicle routing problems tend to emphasize warehouse-to-warehouse or long-haul deliveries, minimizing cost is the main objective for the problem. However to deliver goods in an intra-city area, manual work is required for delivery workers who might have to manually unload goods from the vehicle and take them to the stockroom at customer locations. The vehicle routing problem with manual materials handling (VRPMMH) is introduced as one of classifications of VRP. The delivery activities require workers to expend certain amounts of physical energy. Narrow streets in the city also force supplies or third-party logistics providers to use small vans or trucks instead of trailer trucks to deliver the supplies. It is also common to deliver supplies to several places within one working day. In addition, the unloading time for unloading goods must be included in the working time. Consumer products are just some examples of supplies which have to be delivered to convenience stores, grocery stores, or restaurants that are scattered around the city. These stores do not have large warehouse or storage facility to store their supplies for weeks or months. Usually, they rely on receiving their supplies every day or every fixed interval. Two mathematical models of VRPMMH are developed. Based on additional unloading time and energy expenditure constraints, the objective is to determine optimal fleet size and delivery routes such that the total cost is minimized. The heuristic procedure and the hybrid genetic algorithm are also developed as the approximation approaches to solve for problems in general. Not only energy expenditure of workers is concerned to prevent overload of energy capacity of workers but also workload balancing must be investigated for the equity issue. Workforce assignment model is the second model developed. After the optimal fleet size and delivery routes have been determined from the first model, workers can be selected to the delivery vehicles. The model is intended to assign delivery workers to vehicles to minimize the differences in physical workload. The computational experiments show the significant results for both models. In the first model, test problems are examined and compared among optimal, heuristic, and hybrid GA solutions. The results show that the solutions from approximation procedures are relatively good when comparing to the optimality. For the second model, when comparing the results from the test problems between pre-assigned workforce and worker-vehicle assignment, physical workload balancing show the obviously different results. It is also noted that ILOG CPLEX is adopted to find the optimal solutions for those models

Thammasat University. Thammasat University Library

Address: BANGKOK

Email: preserv@tu.ac.th

Name: Suebsak Nanthavanij

Role: advisor

Name: Chawalit Jeenanunta

Role: co-advisor

Created: 2016

Modified: 2564-07-20

Issued: 2021-07-20

วิทยานิพนธ์/Thesis

application/pdf

DOI: 10.14457/TU.the.2016.1421

eng

DegreeName: Doctor of Philosophy

Descipline: Technology

Grantor: Thammasat University

Level: Doctoral Degree

©copyrights Thammasat University

ลำดับที่.	ชื่อแฟ้มข้อมูล	ขนาดแฟ้มข้อมูล	จำนวนเข้าถึง	วัน-เวลาเข้าถึงล่าสุด
1	3342prachya.pdf	861 KB	1	2023-12-15 16:16:37

ใช้เวลา

0.036508 วินาที