Integration of Historical Data and Ant Colony Optimization to the Vehicle Route Planning – A Case of K Logistics Company
碩士 === 中華大學 === 運輸科技與物流管理學系 === 107 === This study uses the company's own long-term storage of information, such as tracking data, vehicle location data. With reference to the concept of pheromones update in the ant colony optimization, the driver's accumulated vehicle information is retai...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | zh-TW |
Published: |
2019
|
Online Access: | http://ndltd.ncl.edu.tw/handle/cyqrr6 |
id |
ndltd-TW-107CHPI0425003 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-107CHPI04250032019-07-16T03:45:02Z http://ndltd.ncl.edu.tw/handle/cyqrr6 Integration of Historical Data and Ant Colony Optimization to the Vehicle Route Planning – A Case of K Logistics Company 整合歷史資料與蟻群演算法於車輛路線規劃之研究—以K物流公司為例 CHIU,CHIUNG-TA 邱炯達 碩士 中華大學 運輸科技與物流管理學系 107 This study uses the company's own long-term storage of information, such as tracking data, vehicle location data. With reference to the concept of pheromones update in the ant colony optimization, the driver's accumulated vehicle information is retain as expert experience, and the base of the company's knowledge base is established for reference when the company plans the delivery route. Not only can the driver's daily delivery route habits be satisfied, but also the driving route can be stored. It can reduce the adaptation time of new drivers to different areas, and improve the delivery efficiency and reduce the driver's learning cost. In order to construct effective and high-liberal characteristics, this research mainly uses QGIS, Python, MongoDB and other open source free software for development and design. Such will improve the flexibility of adjustment required by enterprises in the future. This study proposed a meta-heuristic method, named as the Historical Data Oriented Ant Colony Optimization (HDOACO), which can effectively integrate vehicle position data and establish a historical pheromone database. The HDOACO method includes this pheromone parameter into the path cost. At the same time, the HDOACO method considers two important factors of driving experience and driving distance in its formula. To confirm the possibility of the proposed HDOACO algorithm, 30 sets of actual cases were tested. After comparing the solved path results with the actual driver's driving mileage, the operating path distance can reduce by an average of 45.44%. Such results imply that the HDOACO method could enhance the new drivers’ efficiency and reduce delivery time. CHO, YUH-JEN SU, JAU-MING 卓裕仁 蘇昭銘 2019 學位論文 ; thesis 44 zh-TW |
collection |
NDLTD |
language |
zh-TW |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 中華大學 === 運輸科技與物流管理學系 === 107 === This study uses the company's own long-term storage of information, such as tracking data, vehicle location data. With reference to the concept of pheromones update in the ant colony optimization, the driver's accumulated vehicle information is retain as expert experience, and the base of the company's knowledge base is established for reference when the company plans the delivery route. Not only can the driver's daily delivery route habits be satisfied, but also the driving route can be stored. It can reduce the adaptation time of new drivers to different areas, and improve the delivery efficiency and reduce the driver's learning cost. In order to construct effective and high-liberal characteristics, this research mainly uses QGIS, Python, MongoDB and other open source free software for development and design. Such will improve the flexibility of adjustment required by enterprises in the future.
This study proposed a meta-heuristic method, named as the Historical Data Oriented Ant Colony Optimization (HDOACO), which can effectively integrate vehicle position data and establish a historical pheromone database. The HDOACO method includes this pheromone parameter into the path cost. At the same time, the HDOACO method considers two important factors of driving experience and driving distance in its formula. To confirm the possibility of the proposed HDOACO algorithm, 30 sets of actual cases were tested. After comparing the solved path results with the actual driver's driving mileage, the operating path distance can reduce by an average of 45.44%. Such results imply that the HDOACO method could enhance the new drivers’ efficiency and reduce delivery time.
|
author2 |
CHO, YUH-JEN |
author_facet |
CHO, YUH-JEN CHIU,CHIUNG-TA 邱炯達 |
author |
CHIU,CHIUNG-TA 邱炯達 |
spellingShingle |
CHIU,CHIUNG-TA 邱炯達 Integration of Historical Data and Ant Colony Optimization to the Vehicle Route Planning – A Case of K Logistics Company |
author_sort |
CHIU,CHIUNG-TA |
title |
Integration of Historical Data and Ant Colony Optimization to the Vehicle Route Planning – A Case of K Logistics Company |
title_short |
Integration of Historical Data and Ant Colony Optimization to the Vehicle Route Planning – A Case of K Logistics Company |
title_full |
Integration of Historical Data and Ant Colony Optimization to the Vehicle Route Planning – A Case of K Logistics Company |
title_fullStr |
Integration of Historical Data and Ant Colony Optimization to the Vehicle Route Planning – A Case of K Logistics Company |
title_full_unstemmed |
Integration of Historical Data and Ant Colony Optimization to the Vehicle Route Planning – A Case of K Logistics Company |
title_sort |
integration of historical data and ant colony optimization to the vehicle route planning – a case of k logistics company |
publishDate |
2019 |
url |
http://ndltd.ncl.edu.tw/handle/cyqrr6 |
work_keys_str_mv |
AT chiuchiungta integrationofhistoricaldataandantcolonyoptimizationtothevehiclerouteplanningacaseofklogisticscompany AT qiūjiǒngdá integrationofhistoricaldataandantcolonyoptimizationtothevehiclerouteplanningacaseofklogisticscompany AT chiuchiungta zhěnghélìshǐzīliàoyǔyǐqúnyǎnsuànfǎyúchēliànglùxiànguīhuàzhīyánjiūyǐkwùliúgōngsīwèilì AT qiūjiǒngdá zhěnghélìshǐzīliàoyǔyǐqúnyǎnsuànfǎyúchēliànglùxiànguīhuàzhīyánjiūyǐkwùliúgōngsīwèilì |
_version_ |
1719223717653905408 |