Rapid humanitarian assessments and rationality: a value-of-information study from Iraq, 2003-04

Rapid assessments are one of the standard informational tools in humanitarian response and are supposed to contribute to rational decision-making.(1) The extent to which the assessment organisation itself behaves rationally, however, is an open question. This can be evaluated against multiple criteria, such as the cost and value of the information it collects and its ability to adapt flexibly design or samples when the survey environment changes unforeseeably. An unusual data constellation from two concurrent recent (2003-04) rapid assessments in northern Iraq permits us to model part of the actual assessment behaviour in terms of geographical, community and prior substantive information attributes. The model correctly predicts the decisions, in 79 per cent of the 2,425 local communities in focus, that data collector teams in the Emergency Mine Action Survey made to visit or not to visit. The analysis demonstrates variably rational behaviour under conditions of insecurity, repeated regrouping and incomplete sampling frames. A pronounced bias towards very small rural settlements is irrational for the overall results, but may be a rational strategy of individual survey workers seeking to prolong their employment. Implications for future assessments are sketched in the areas of tools for urban surveys, greater adaptability, including early feedback from users, and sensibility to value-of-information concepts.     

城市震后救灾系统救灾决策研究

论述了震后救灾系统模型及评估方法,采用Ｍｏｎｔｅ Ｃａｒｌｏ模拟和最优化技术理论,建立了城市震后救灾、救护、运输网络系统的可靠性分析方法和以失效概率为控制参数的最佳路径优化方法,并以某城市网络系统为例通过分析研究,给出了震后救灾决策方法。     

淮南市防震减灾决策系统的建立

在机制上将一个地市(淮南市)的地震预报与防震决策融为一个动态整体,从而建立起一个实用性强的防震减灾决策系统,使城市各种防震对策的生成有了更科学的基础。成为市政府在防震减灾决策中不可缺少的工具。     

基于粗糙集理论的路段交通事故多发点成因分析

交通事故多为多种原因综合造成且具有不确定性。判别各因素对于引发交通事故的影响程度存在困难,因此,需要引入不确定性分析方法。基于粗糙集理论具有处理不精确、不确定与不完全数据的优势,是一种先进的并处于不断发展的不确定性分析和推理方法。笔者提出路段上交通事故成因分析的模型和方法。具体做法是,根据该路段事故多发点的统计数据建立基于粗糙集理论的决策表,再利用粗糙集模型的简约算法求出各个因素的重要性,从而判断各因素对交通事故的影响程度大小,为决策提供依据。以算例说明模型和方法的可用性。     

我国公众参与环境决策的运行机制及其优化

本文简要论述了良好的决策机制是公众参与环境决策的基本保障,分析了当前我国公众参与环境决策运行机制存在的不足,并提出了优化其运行机制的一些思路。     

多目标排队决策法——在大连地区土地利用结构调整中的应用

本文采用多目标排队决策法对专家的咨询意见进行了定量处理,从各用地部门的经济、生态和社会效益出发,确定了他们的优先级,并结合土地利用现状对今后的土地利用结构作出了合理的调整规划设想。     

公路隧道信息化施工与计算机辅助决策系统研究

结合公路隧道建设实例,通过需求分析,提出了公路隧道信息化施工与计算机辅助决策系统的总体思路。利用数据库技术与软件工程技术,对此系统进行了详细设计,并开发了此软件系(Tunnel-CAS)。该软件系统集围岩类别智能判别、围岩应力分析与预测、围岩变形分析与预测和围岩支护结构安全性判别四大功能于一体,具有友好的用户界面。     

Multicriteria decision support in a technology competition for cleaning polluted soil in Helsinki

Due to expansion of the capital area in Finland, industrial areas are being replanned for residential and commercial use. The soil in these areas is sometimes contaminated, and must therefore be cleaned before building. In spring 1997 the City of Helsinki and the National Technology Agency of Finland declared a contract-based competition for cleaning the polluted soil of the planned Toukolanranta residential area. Nine proposals entered the competition, and the problem was to choose three best candidates for test-cleaning a small part of the region considered. The winner of the test-cleaning phase will get the contract for cleaning the whole area. The proposals were evaluated based on five criteria defined by the competition board consisting of eight experts. The finalists were chosen aided by the SMAA-2 Stochastic Multicriteria Acceptability Analysis technique. SMAA-2 is a method for analysing what kind of preferences favour each alternative. This method is particularly useful in applications, where it is difficult or impossible to obtain accurate preference information. In this application SMAA-2 was observed to suit well for choosing a small set of best alternatives with different strengths and weaknesses.