Publications

Wu, B., Yan, X.P., Wang, Y., Zhang, D. and Guedes Soares, C. (2017), “Three-Stage Decision-Making Model under Restricted Conditions for Emergency Response to Ships Not under Control”, Risk Analysis: An International Journal, Vol. 37(12), pp. 2455-2474

This paper develops a three-stage decision-making model for emergency response to not under command (NUC) ships, a typical maritime incident worldwide. After summarizing the features of emergency response in the introductory section, emergency response to NUC ships are investigated in Yangtze River from 2007 to 2012. The detailed causation distribution and traditional actions of NUC ships are analyzed by using the statistical data. Moreover, A three-stage decision-making framework, aims to make quick and effective response to NUC ships, are proposed to meet its distinguishing requirements, including time limitation, resource constraint, and information asymmetry. First, emergent situation is identified by considering resources constraint in limited time. Ship condition, navigational environment, and available rescue resources, are assumed to be the influencing and restricted resources of feasible actions to NUC ships. Second, the restricted conditions are quantified as interval numbers according to the statistic data, and a case based reasoning (CBR) method is introduced by treating the restricted conditions as attributes. Since the CBR method is challenged for incomplete information of the historical data, it is only used to select the basic feasible options according to the similarity of the new case and existing cases. Last, the final decision is made by using a single Bayesian network according to expert judgments. Moreover, as the participated disciplines, NUC ship, ships passing by, and maritime safety administration (MSA), may rely on the actions of former decision-making disciplines. A merging Bayesian network, compared to the single Bayesian network, is developed to address this information asymmetry problem. Specifically, the conditional probability tables (CPTs) under the decisions of former discipline, carried out by reconsidering the expert judgments on its own options, are incorporated into the single Bayesian network to link them together as a merging Bayesian network. This improved Bayesian network can make a comprehensive decision to facilitate the emergency response with better accuracy. Moreover, a liner programming model is developed to analyze the information asymmetry in decision-making, and the result can assist the decision-maker to make correct decisions with incomplete information.

If you did not manage to obtain a copy of this paper: Request a copy of this article



For information about all CENTEC publications you can download: Download the Complete List of CENTEC Publications