整合型軌道提速最佳化決策支援模式之研發

Abstract

現今軌道運輸多朝向高速化發展，然而軌道所費不貲，如何有效運用現有資源，達成軌道提速的最大效益，成為一項重要的課題。目前軌道提速可分為(1)改善基礎設施(2)提升列車技術兩種方法。前者以改善軌道基礎設施，使列車能以更高的營運時速行駛於改善路段上；後者則是提升列車的性能，使列車能克服原有路段設施的限制，以更高的速度行駛。而軌道營運條件的改善，對列車的服務設計即列車發車頻率與停站模式勢必有所影響，換言之軌道提速與服務設計兩者有密切的連動關係。然而過往研究泰半單獨討論軌道提速中「改善基礎設施」方法，鮮少涉及列車技術的層面；過往服務設計的研究中，也少有觸及軌道提速後相應的營運方式。是以本研究希冀建立一「整合型軌道提速最佳化決策支援模式」，能統合處理軌道提速及服務設計。本系統透過營運業者輸入必須的基本資料後，能由內部建立之「軌道提速方案產生器」中產生出可行之軌道提速方案，經由研發的最佳化模式判別其中最合適之方案，並因應營運業者之需求，建立對應之列車服務設計，最後完成軌道提速及服務設計之整合決策。因而本決策能提升整體軌道資源運用效率，提升整體軌道投資績效。 本研究發展「整合型軌道提速最佳化決策支援模式」中包含一組整合型軌道提速最佳化決策支援模式，以及協助求解的加速求解演算法，加速求解演算法之運作，能有效提升演算問題之求解規模及效率。本研究設計兩組案例，其一為虛擬案例，用以證明使用加速求解演算法後，求解之準確性及效率；第二組案例以北宜直線鐵路為對象進行分析，藉由顧問公司取得之原始資料，得出以南港、頭城之直線鐵路為最佳方案，並以TEMU為最佳車種方案。

Incremental improvement is the most common, near-term approach to upgrading railway corridors to achieve high speeds. However, investing in incremental improvement is expensive, efficient utilization of the investing is one of the objectives. Existing infrastructure or rolling stock can be improved in various ways to allow for increasing speeds and reducing travel time along a route. Each track section has a characteristic set of opportunities for increasing speed and their corresponding effect on travel time reduction, along with an associated set of costs. Similarly, each type of rolling stock has the potential to increase operational speed and reduce travel time, corresponding to a specific price tag. And investing in both existing infrastructure and rolling stock should consider the Life-cycle cost （LCC）. This research focuses on the development of a decision support process, Railway Route Improvement Optimizer (RRIO), using mathematical programming to identify the most cost-effective strategy for reducing corridor travel time given a prescribed performance goal and budget. Based on information of the current network and traffic, and the available investment options, numerical results show that RRIO can successfully determine the optimal solution to which sections should be upgraded, the type of rolling stock that should be implemented, and the fleet size. To improve the solution efficiency, we also developed a solution algorithm to improve the solution efficiency by nearly 85% and maintain the solution quality within 3% optimality gap. With this tool, railway agencies can maximize return from capital investment and maintain reliable services at the same time.