以群組為基礎之幹道號誌連鎖──調和複雜的輕軌平面交岔與小客車續進

Abstract

隨著都市交通需求日趨多樣與複雜，輕軌系統已成為現代城市中不可或缺的大眾運輸系統之一，其運行效率與一般車流的協調日益受到重視。針對台灣都市輕軌系統多採B型路權、平面交岔複雜且與一般車流高度互動的特性，本研究考量轉向車流、路口型態變化與輕軌動線調整，並提出「以群組為基礎之幹道號誌連鎖」的被動式大眾運輸優先號誌策略。研究發展混合整數線性規劃(MILP)模型涵蓋兩階段：第一階段針對各路口獨立最佳化號誌時制，第二階段則進行路網連鎖號誌最大化帶寬設計，兼顧輕軌與一般車流效率。透過三組數值實驗，系統性檢驗模型於不同交通需求與目標導向下的適應性與效能。此外，結合交通模擬軟體進行評估，進一步驗證所提模型在實際交通環境中的應用潛力。以一案例分析，若強調純輕軌優先，則比小客車為主之情況，可使輕軌延滯降低10.8%，但車輛則減少了6.8%的續進效率。而本研究均衡輕軌與一般車輛，雖使輕軌延滯增加3.5%，略高於輕軌優先，但車輛續進效率提升6.4%。透過數值分析及模擬驗證，結果顯示適當調整多目標權重參數，能在提升輕軌運輸效益的同時維持一般車流順暢，展現模型可應用於複雜都市路網。

As urban transportation demands become increasingly diverse and complex, light rail transit (LRT) systems have become indispensable components of modern cities, with growing emphasis on the coordination between LRT operations and general traffic flow. Considering the characteristics of Taiwan’s urban LRT systems—predominantly Category-B right-of-way, complex at-grade intersections, and high interaction with general traffic—this study incorporates turning movements, intersection geometry, and LRT alignment adjustments to construct a group-based arterial signal coordination strategy employing passive transit signal priority (TSP). This study has developed a mixed-integer linear programming (MILP) model with two stages: the first stage independently optimizes signal timing at each intersection, while the second stage maximizes progression bandwidth across the network to balance the efficiency of both LRT and general traffic. Three sets of numerical experiments—baseline model operation, traffic flow volume analysis, and vehicle-type weighting analysis—systematically evaluate the model’s adaptability to various traffic demands and objectives. To further validate the proposed model and assess the model applicability, this study applies microscopic simulations under a real-world tram network. It can be shown that prioritizing LRT over vehicle-priority reduces LRT delay by 10.8%, but results in a 6.8% decrease in vehicle progression efficiency. By contrast, the balanced approach adopted in this study—considering both LRT and general vehicles—leads to a 3.5% increase in LRT delay (slightly higher than LRT priority) but improves vehicle progression efficiency by 6.4%. Numerical analysis and simulation validation demonstrate that appropriately adjusting multi-objective weight parameters can enhance LRT operational benefits while maintaining smooth traffic flow, highlighting the model’s applicability to complex urban networks.