Optimization of Numbers of Self-Run and Outsourced Trips for Home Delivery Carrier Considering Stochastic Demands and Feasibility

View Statistics Email Alert RSS Feed

Information

Details

Project title

Code/計畫編號

MOST106-2410-H019-021-MY2

Translated Name/計畫中文名

考量隨機需求與可行性下宅配業自有與外包趟次最佳化之研究

Project Coordinator/計畫主持人

Ching-Hui Tang

Funding Organization/主管機關

National Science and Technology Council

Department/Unit

Department of Transportation Science

Website

https://www.grb.gov.tw/search/planDetail?id=12284547

Year

2017

Start date/計畫起

01-08-2017

Expected Completion/計畫迄

31-07-2018

Bugetid/研究經費

755千元

ResearchField/研究領域

經濟學

"本研究擬針對宅配業轉運中心自有與外包趟次數量問題進行探討，首先考量營運中貨物需求之隨機性，並同時考慮業者在實務上每月預先與外包業者簽訂之月趟次以及即時營運產生之臨時趟次，以規劃一符合隨機需求擾動之自有與外包趟次。另外，過去研究大多基於求解結果為可行之基礎上，構建數學模式以求得最佳解，然而此方式忽略了最佳解於實際營運時之不可行性，所求得之最佳解可能過於「樂觀」，進而增加其執行之困難度。有鑑於此，本研究考量業者自有駕駛臨時出車之機率，並進一步考慮自有臨時趟次數量之可行性，以規劃一滿足可行機率下的自有與外包趟次，增加其在實際執行時的可行性。本研究擬利用數學規劃方法結合統計學中機率分配概念，分別構建隨機需求模式與可行性隨機需求模式。其中，在可行性隨機需求模式中，除考量貨物需求隨機性外，在模式中加入自有車臨時趟次之機率分配，以表達自有車總臨時趟次之機率與可行性。另外，本研究針對兩模式推導出相關特性，並運用所推導之特性、兩階段隨機模式特性、分解式概念與基因演算法中的尋優機制，設計一反覆迭代兩階段子問題之求解演算法。最後，我們擬以國內一宅配業者資料為例進行分析，並根據研究的結果，提出結論與建議。""In this research, we will deal with the problem of the numbers of self-run and outsourced trips for a transshipment center for a home delivery carrier. Stochastic demands which are usually occurred in actual operations are considered. We will simultaneously consider monthly trips which are contracted in advance in monthly agreements with the outsourcing carrier and temporary trips which are occurred in real-time operations to plan the numbers of self-run and outsourced trips under stochastic demand disturbances. In addition, most past studies developed a mathematical model to find an optimal solution based on that the obtained optimal solution is applicable in real world operations. However, the infeasibility of the optimal solution was neglected and thus the obtained optimal solution may be over optimistic, making it difficult to be applied in actual operations. Therefore, we will consider the probability of the self-run driver for serving a temporary trip and further involve the feasibility of the self-run temporary trip in actual operations. It is expected to plan the numbers of self-run and outsourced trips to meet the feasible probability which improves the feasibility in actual operations. We will use the mathematical programming method combined with the probability distribution concept in Statistics to develop a stochastic demand model and a feasibility and stochastic demand model. In the feasibility and stochastic demand model, we not only consider stochastic demands, but also use a probability distribution of the self-run driver for the temporary trip to reflect the probability and feasibility of the total number of the self-run temporary trip in actual operations. In addition, we will derive several properties from the two models and apply the two-stage stochastic model properties, the decomposition concept, and the searching mechanism in Genetic Algorithm to develop an iterated two-stage sub-problems algorithm. We will perform a case study using the real operating data from a home delivery carrier in Taiwan. Finally, conclusions and suggestions will be given."

Keyword(s)

宅配業
自有與外包趟次
隨機需求
可行性
分解式概念
兩階段隨機模式
求解演算法
Home delivery carrier
Self-run and outsourced trips
Stochastic demands
Feasibility
Decomposition concept
Two-stage stochastic model
Solution algorithm

DSpace CRIS

Optimization of Numbers of Self-Run and Outsourced Trips for Home Delivery Carrier Considering Stochastic Demands and Feasibility

Details

Description