緊急救災物流輸配送系統模式構建

Abstract

日常生活中的天然災害，如地震、颱風、水災、乾旱等，經常造成人們生命與財產的重大損失。由於資源的質與量皆為有限，危難管理者(emergency managers)必須就有限資源，規劃最佳的運用時程與策略，以在最短時間內，將救援物資緊急配送至災區。而災害救援行動中物流輸配(disaster relief management)的基本問題，係在考量時間因素的情況下，如何使用不同的運輸模式，透過不同的運輸網絡，將各種救援物資從各物資集散中心，順利的配送至災區。其中，如何有效地分配與調度物資，以避免浪費或堆置無用，將是決定此配送系統優劣的關鍵。 本研究所構建之「救災物流輸配送系統」包括前端「物資指派」及後端「物資配送」兩個部份，物資指派部分在於如何將所有供給端所提供之物資指派至各區域型配送中心。第二部分探討資源指派並探討當災害發生初期所衍生之供需失衡問題，以數學方法計算出各災區對於物資配送之權重及其配送之優先順序。同時我們也採用互動式多階層規劃法在將問題分為兩個子問題在決策與運作兩層級中其間有交互影響之關係下游每階段之運作結果將回傳至上游以提供下一時階之規劃，進而提升整體系統之最佳化。

Natural disasters which might include earthquakes, typhoons, floods, drought, etc. are part of our life. They have significant devastating effects in terms of human injuries and property damage. Because quantity and quality of the resource are limiting factors, emergency managers do have to find an optimal schedule for assigning resources in time to the affected areas, and the basic underlying logistical problem for disaster relief management is to move a number of different commodities using a number of transportation, from a number of distribution center to one or more affected area over a transportation network in a timely manner ,and how to distribute relief effectively and efficiently is one of the key point. In this study will propose a relief distribution system decompose into two parts, concentrating goods and materials and dynamic resource allocation, the first part is how to concentrate goods and materials in the shortest time and minimize the fleet size. And at the second part we discuss when concentrated goods and materials are insufficient for demand of affected areas we use fuzzy clustering to obtain the weight of affected area. Simultaneously, we will use interactive multi-level programming method because the decisions inherent in the problem decompose hierarchically into two sub-problems where tactical decisions are made at the top level, and the operation distribution and evacuation are made at the base level. Consistency between the decomposed problems is achieved with an interactive coordination procedure which transfers anticipated information form the base level to improve the top level decisions to optimal performance of the relief distribution system.