Information System (Service) Design Based On Customer Satisfaction

With the degradation of the global economic environment, the competition of themarket is becoming more intense. Information interconnections in cross-organizations,business process reengineering and individuation present new challenges toinformation system design. Reusable concepts and methods give directions to solvethis problem through producing or releasing reusable components (Web Services) andrapid constructing information systems through component combinations. These cannot only improve the satisfaction of the customer for the information system, but alsocan effectively reduce the cost of development and application of the informationsystem, and make it responds quickly to the market.Currently popular method of Service Oriented Architecture (SOA) and traditionalmodel of Component-based Software Development (CBSD) all require a completeclient-oriented framework model to support Web services or model reuse. This paperconstructs a customer-oriented information systems architecture design frameworkmodel. This model includes the expression, analysis, classification of customerdemands and extracts key demands. Then, Quality Function Deployment is used tointegrate key demands into reusable components. Finally, a multi-objectiveoptimization method is proposed for information system design to reach customersatisfaction.The main work of this paper include:1) The demands expression and demandsmodeling: we propose a descriptive model of customer demands as four-tuple incustomer demands analysis phase. The expression of the customer demands in tuple isfacilitated to compare.2) Demands package: In the customer demands modeling stage,the Quality Function Deployment (QFD) method is used combining evolutionaryalgorithm (NSGA-Ⅱ). It also includes extract key demands and optimizes the size ofdemands module package.3) Design optimization: a multi-objective optimizationmodel based on customer satisfaction. Customer Satisfaction is consisted of functionsatisfaction and quality satisfaction. Function satisfaction and quality satisfaction alsohave an impact on cost estimation. Under cost constraints, a multi-objectiveoptimization model as the basis for the design is proposed using binary particle swarm algorithm to solve the problem. The final experiments of the two examples in differentscenarios show that the efficiency of multi-objective optimization methods. Theobtained Pareto frontier can provide richer information for decision-makers to meetthe needs of different customers at different levels.In this paper, information system design methods are to encapsulate demandsmodules or services as the basic unit and to use the customer satisfaction optimizationas objectives. With the combination of modules could meet different customer needs.This method provides theories and methods for the next generation of SOAapplications and information system design.