考量消防應變效率之狹小巷道改善策略：
基於都市路網分析之觀點

狹小巷道是造成消防延誤的潛在因素，不論是巷道的寬度不足或是違停車輛的停放都可能阻礙消防車的通行，因而延誤救災時機。由於過往學界均未有發展以網路設計提升消防效率之研究，因此本論文旨在提供一個路網方法，以輔助狹小巷道改善之決策，其中包括（1）劃設消防通道（2）拓寬狹小巷道。首先，以路網的Voronoi圖劃分巷道火災風險負責範圍，再評估建築風險以計算個別巷道的火災風險。考量到消防單位在巷道中的移動限制，本研究建構了雙層隨機阻塞路網來描述消防救災之行動，並藉由分割街廓子路網的方式，將替代路徑的搜索範圍加以限縮。接著，利用蒙地卡羅方法估計出巷道旅行時間之期望值，以此作為消防服務可及性的指標。在巷道保障通行之關鍵性上，逐一評估巷道通行受到確保之下所增進的消防可及性，再利用關鍵性之排序作為消防通道劃設之參考。而在巷道拓寬的部分，首先分別列舉出個別街廓的巷道拓寬計畫，並計算其效益與成本，再以背包問題決定出限制成本下的巷道拓寬計畫。在案例分析中，我們將研究方法應用於台北市西區舊市街。結果顯示，本模型能夠辨識保障通行之最關鍵巷道，另外也能求得最有利的巷道拓寬決策，從而促進狹小巷道社區之消防可及性。本研究建議後續研究者進一步探索巷道阻塞之特性，或考慮消防滅火與水源中繼之因素，藉此進一步精進模型的建立。

關鍵字

狹小巷道；消防可及性；隨機阻塞；巷道改善；關鍵性；蒙地卡羅方法；沃羅諾伊圖

並列摘要

Narrow Alleys are the potential cause of delays in firefighting operations. Once a fire breaks out in an old urban community, fire engines may be blocked in an alley due to insufficient width or illegally parked vehicles, thereby delay firefighting operation. However, few existing studies have assessed the threat of narrow alleys and improvement of firefighting in a systematic manner. Hence, this thesis proposes a network-based method to support decision-making in alley improvement measures, which include (1) marking fire lanes and (2) widening alleys. First, the Voronoi diagram of the road network is used to divide alley-based zone, and then the fire risk of each alley is estimate by summing up the fire risk of buildings within the alley. Considering the mobility of fire crews in alleys, we construct a two-level road network with stochastic alley blockage to describe pathfinding in firefighting. Also, the division of block subnetworks limits the searching space of alternative paths. Then, the Monte Carlo method is used to estimate the expected travel time over targeted alleys to evaluate fire service accessibility on the network. To assess the criticality based on passability assurance of alleys, a full-scan approach is conducted to evaluate the improved accessibility index after ensuring their passability. The ranking of the alley criticality provides a reference to the priority of setting fire lanes. To determine the widening alley, this study proposed a block-based widening plan generation process and evaluate the benefit and cost of each widening plan. Then, a knapsack problem is utilized to determine the widening decision over candidate alleys with a budget constraint. In the case analysis, we apply the solution method to the western of Taipei City. It is demonstrated that this model is able to identify the most critical alleys for passability assurance and determines the most cost-effective alley widening decision to enhance the fire service accessibility in the communities. For future research, the nature of alley blockage can be further explored and fire extinguishing or water transport can be further considered to elaborate on the model and analysis.