An Expert System Based on Multi-reasoning Mechanism for Port Machine Diagnosis

Expert system plays an important role in port machine diagnosis, which aims at automatic equipment test for higher availability and efficiency of port operations. In this study, a port machine diagnosis expert system is proposed based on multi-reasoning mechanism. Relying on the knowledge acquired from the experienced experts in the port machine engineering, the system builds a library of relative experience and a set of rules of reasoning and estimating. Multi-reasoning mechanism that simulates the decision-making process of domain experts is employed to achieve reliable diagnosis results. The reasoning machine integrates artificial neural network, uncertain decision making and decision tree, which complements each other by sustainable growing voting mechanism. The effect of this multi-reasoning mechanism is evaluated and validated by means of Matthew＇s Correlation Coefficient （MCC）. The system incorporating the mechanism is successfully designed, implemented and applied in Shanghai Port.

Uncertain Dispatching Decision and Genetic Algorithms for Railyards Operations

In this paper, a fuzzy dispatching model has been developed and genetic algorithms have been used for considering the corrdination among uncertain decisions in railyards dispatching plan. The objective considered here is to maximize the average grade of satisfaction including the yards output and ontime service of the plan. Based on the analysis of the fuzzy scheduled model for the problem, the chromosome and fitness function of genetic algorithms are proposed. The genetic operators are designed and the genetic algorithms are given. Experimental results show that the genetic algorithms and fuzziness approach could be a promising way for railyards dispatching problem.

Man-machine Function Allocation Based on Uncertain Linguistic Multiple Attribute Decision Making

Function allocation is one of the necessary stages in the design course of man-machine systems since appropriate function allocation makes the whole system more effective, reliable and inexpensive. Therefore, our research mainly focuses on the problems of function allocation between man and machine in man-machine systems, analyses each capability advantage of man and machine according to their respective inherent characteristics and makes a comparison between them. In view of highly uncertain characteristics of decision attribute value in the practical process, we introduce the uncertain linguistic multiple attribute decision making （ULMADM） method in the function allocation process. Meanwhile, we also use the uncertain extended weighted arithmetic averaging （UEWAA） method to determine the automation level range of the operator functions. Then, we eventually estab- lish the automation level of man-machine function allocation by using the multi-attribute decision making algorithm, which is combined by UEWAA and uncertain linguistic hybrid aggregation （ULHA） operators. Finally, an example about function allocation is given, that is, fault diagnosis in the cockpit of civil aircraft. The final result of the example demonstrates that the proposed method about function allocation is feasible and effective.