【正文】 ement. Prentice Hall,1999 [14](英),: [15]ChenTung Fuzzy Approach to Select the Location of the Distribution Sets and Systems,2001 [16] Size and Location Planning of Public Logistics Research Part E35(1999) [17]?；鄱?2000(3) [18]王戰(zhàn)權(quán),2000(1) [19]?；鄱?2001(2) 報,2001(9) 特區(qū)理論與實踐,2000(4) 護,2001(6) [20]王軍鋒,馮平,——,2001(2) [21]、,2002(1) [22]李銘,2002(3) [23] ,2002(4) [24], [25],2001(5) [26], [27], [28],1997(16) [：廣東旅游出社,2000 [30]賈正稅,潘洪達,金鳳閣,2000(3) [31]余朵茍 [J].物流科技2008(11):15.[32] 孟參 [J].物流技2006(9):1517.[33] 馮旭東 [J].中國物流與采購 2003(23):31.[34] [D].哈爾濱 哈爾濱工業(yè)大學(xué) 2006.[35]李文峰，袁兵，張煜編著物流系統(tǒng)建模與仿真。北京：科學(xué)出版社，2010。[36]張曉萍主審，彭揚 五承健 編著物流系統(tǒng)建模與仿真 浙江大學(xué)出版社，2009[37] Robust optimization for emergency logistics planning: Risk mitigation in humanitarian relief supply chainsABSTRACT：This paper proposes a methodology to generate a robust logistics plan that can mitigate demand uncertainty in humanitarian relief supply specifically, we apply robust optimization (RO) for dynamically assigning emergency response and evacuation traffic flow problems with time dependent demand paper studies a Cell Transmission Model (CTM) based system optimum dynamic traffic assignment adopt a min–max criterion and apply an extension of the RO method adjusted to dynamic optimization problems, an affinely adjustable robust counterpart (AARC) experiments show that the AARC solution provides excellent results when pared to deterministic solution and sampling based stochastic programming insights of RO and transportation that may have wider applicability in humanitarian relief supply chains are provided.Introduction Over the past three decades,the number of reported disasters have risen , 5 billion people have been affected by disasters with an estimated damages of about trillion dollars Although most of these disasters could not have been avoided, significant improvements in death counts and reported property losses could have been made by efficient distribution of supplies here could mean personnel, medicine and food which are critical in emergency supply chains involved in providing emergency services in the wake of a disaster are referred to as humanitarian relief supply relief supply chains are formed within short time period after a disaster with the government and the NGO’s being the major drivers of the supply chain. Clearly, emergency logistics is an important ponent of humanitarian relief supply chains. Most literature in emergency logistics focuses on generating transportation plans for rapid dissemination of medical supplies inbound to the disaster hit regionThere is, however, another aspect of emergency logistics which is often ignored – outbound logistics. The outbound logistics considers a situation where people and emergency supplies (. medical facilities and services for special need evacuees) need to be sent from a particular location affected by disaster within a given time horizon. In the outbound emergency logistics, the demand of traffic flows is usually highly uncertain and depends on a number offactors including the nature of disaster (natural/manmade) and time of uncertainty in the demand causes disruptions in emergency logistics and hence disruptions in humanitarian relief supply chains leading to severe suboptimality or even infeasibility which may ultimately lead to loss of life and order to mitigate the risk of uncertain demand, we study the problem of generating evacuation transportation plans which are robust to uncertainty in outgoing demand. More specifically, we solve a dynamic (multiperiod) emergency response and evacuation traffic assignment problem with uncertain demand at source nodes. Researchers and practitioners in the field of transportation are concerned with multiperiod management problems with an inherent time dependent information dynamic optimization approaches for dealing with uncertainty(. stochastic and dynamic programming) usually require the probability distribution for the underlying uncertain data to obtain expected , in many cases, it may be very difficult to accurately identify the distribution required to solve a , in many cases, it may be very difficult to accurately identify the distribution required to solve a problem. In addition, the robust solution guaranteeing the feasible evacuation plan is important since infeasible solutions may cause the potential loss of life and property in extreme events. We explore the potential of robust optimization (RO) as a general putational approach to manage uncertainty, feasibility, and tractability for plex transportation approach has been originally developed to deal with static problems formulated as linear programming (LP) or conicquadratic problems (CQP), using crude uncertainty with hard means that uncertainty is assumed to reside in an appropriate set and RO guarantees the feasibility of the solution within the prescribed uncertainty set by adopting a min–max RO technique has been successfully applied in some plex and large scale engineering design and optimization problems similar as robust control in control theory. The original RO approach considers static problems. The underlying assumption of RO is “here and now” decisions, and all decision variables need to be determined before any uncertain data are is not typical in many transportation management problems that have the multiperiod nature. In multiperiod transportation problems such as dynamic traffic assignment, “wait and see” decisions are made, which means some decision variables are “adjustable” and affected by part of the realized the need to account for such dynamics, BenTal et al. (2004) have extended the RO approach and developed an affinely adjustable robust counterpart (AARC) approach to consider “wait and see” decisions. To demonstrate the use of AARC to emergency transportation management settings, in this paper we consider a system optimum dynamic traffic assignmen