提升國際快遞業回復力之策略模式

Abstract

人們普遍意識到天然災害和人為災害具有突發性，當擾動發生時，原先的營運計畫可能偏離最佳化的結果甚至導致原先的計畫不能使用，所以調整的手段或重新最佳化的工具是需要的，這些手段或工具可以有效率地使用資源，讓偏離最佳化的營運及時地回到運行的軌道上。 國際快遞業相較於其他運輸服務是一個對時間高度敏感的產業，需要快速的反應供應鏈的擾動，否則可能會失去競爭能力。相較於在擾動發生的階段武斷地做出匆促的決定，本研究提出一個量化的方法，不論在路網上有多少可用的資源或有多少運能可供租借，此方法可根據整合資源分配的概念最佳化回復策略。本研究之問題被模式化為一個多樞紐、多運具、多運輸業者和多商品的路網問題，此量化分析模式可幫助國際快遞業在擾動發生時決定新的替代路線和租賃活動(包括運具的選擇與業者的選擇)。考量到國際快遞業對時間敏感的特性，模式也加入了貨物價值時間函數，讓不同價值的貨物可以以不同方式來運送，獲得較高的顧客滿意程度。 最後，我們執行數值的試驗和敏感度分析。數值試驗的結果證實模式可以應用在實際路網上，並且可提供具實用性的回復策略; 敏感度分析的結果則顯示回復成本的增加對目標式的影響大於運輸時間，而當運輸時間超過某一門檻值後，運輸時間的增加對目標式的影響會急遽上升。

There has been a general awareness that natural and man-made disasters may occur suddenly. When disruptions occur, the previously operational plans may become far from optimal or even infeasible, and means are needed for adjusting or re-optimizing the original plan to adapt the changing environment and to get back on track in a timely manner while effectively using the available resources. International express is one most time-sensitive industry, which may need to respond disruptions quickly so as to improve service quality and to avoid losing their competitiveness with other express service providers. Instead of arbitrarily making rush decisions during the post-disruption phase, this paper contributes a method for quantifying and optimizing the resilience strategies based on an integrated resource assignment concept, regardless of how the available resources are located with respect to the studied logistics network or how many capacities we can rent from others. The studied problem is formulated as a multi-hubs, multi-modes, multi-carriers, and multi-commodities network problem. The analytical model is developed for determining the alternative routes and rent activities (including the mode choice and carrier selection) after the disruption occurs. It also takes into account nonlinear cargo value functions of time to reflect the feature of the express industry that allows company transport different types of cargo with different ways to achieve higher customers’ satisfaction. Numerical experiments are conducted to examine our model applied in more complex networks and real world cases. Through a series of sensitivity analysis, some managerial implications are suggested to decision makers and potential stakeholders.