強震發生後，往往對災區造成相當大的破壞，災區居民常陷入孤立無援的困境，亟需外界給予支援與救助，因此在地震發生後，道路「緊急搶修」為搶修單位第一考量的事項。此外，若交通系統受損，導致救援以及緊急物資車隊無法於第一時間順利進入災區，搶救速度受到牽累，救援工作無法順利進行，民眾之生命與財產可能造成難以估計的損失。 緣此，本研究將針對災後道路緊急搶修問題進行研究與探討，根據其問題特性建構數學模型，並發展出一套建構初始途程與途程改善之兩階段啟發式演算法其中，途程改善以螞蟻族群演算法為主軸，並結合路線間1-1節點交換法，以建構出災區內各災點之最佳搶修順序及各工程隊之搶救排程作業。最後，本研究產生12組測試例題，進行例題測試分析，進而提出結論與建議。 Earthquakes often cause great damage to the transportation infrastructure of the disaster area that isolated the residents of the disaster area. The first priority should be planning emergency repair schedule to repair the damaged transportation infrastructure otherwise food, water, rescue equipment and repair team can not enter the disaster area. Rescuing effort will also be delayed causing immeasurable loss of life and properties. This research studies the repair schedule with minimized total repair time of damaged transportation system after natural disasters. This problem is a variance of standard vehicle routing problem with time windows. A mixed integer optimization model is also built and a meta-heuristics which consists of tour construction and tour improvement procedure is developed. Tour improvement is based on ant colony system combined with 1-1 exchange method to construct the best repair sequence of each disaster node. Finally, twelve test problems are examined for the purpose of this study, and then recommendations are made from the analysis of those test problems.