The Improved Analytic Hierarchy Process And The Implementation Based On B/S Framework

People often face some decision problems and need to make a choice on theseissues. When decision problems people confronted are very complicated, it’sdifficult to make correct choices directly depending on individual subjectivejudgments, so we need to analyze decision problems and provide decision supportinformation to decision-makers by scientific decision-making methods. TheAnalytic Hierarchy Process(AHP) is a scientific decision-making method, whichimplements a combination of qualitative analys is and quantitative analysis and isvery suitable for semi-structured or unstructured decis ion problems. So it not onlyprovides a good support for optima l ordinal arrangement of such decision-makingproblems，but also has the characteristics of simplicity and practicality. But as thewidely used in the field of decision problems, the drawback is gradually revealed.This paper introduces an overview of AHP and improves it in two aspects of AHP.The main contents inc lude three parts as follows.Firstly, this paper introduces an improved method of AHP——Group AnalyticHierarchy Process, analyzes the advantages of GAHP, and proposes an optimizedmethod for inconsistency pair-wise comparison matrix, which input bydecision-makers, to improve the consistency of pair-wise comparison matrix.Secondly, this paper propose a new improved AHP——Cloud Model AnalyticNetwork Process. It incorporates cloud model theory into analytic network processand realizes conversion from quantitative to qualitative in analytic network processthrough reverse cloud generator of cloud model theory. Not only it enriches theanalytic hierarchy process in model structure, but also the integration of the cloudmodel theory is able to build-up decision-making intent of a number ofdecision-makers and reflect the fuzziness and randomness in process of assessment.Finally, these two methods are taken as the ma in evaluation algorithms,supplemented by other several evaluation algorithms, a set of software for theassistance evaluation system based on B/S framework are developed, satisfies users’requirement of evaluation algorithm diversity, and has the characteristics ofdistribution because of the imple mentation of B/S framework; Assisted assessmentevaluation system adopts the browser as the interaction interface, so as to simplifythe client’s requirements, as long as the clients install the IE browser, and within thesame LAN as the server, clients can access the system, provides a simple andconvenient assessment environment for decision-makers; The software uses Oracledatabase to manage the related data, Oracle database ensures that the storage andretrieval of large amounts of data, as well as related data processing, other basic operations and the security of the related data; The implementation of the softwarecan effectively avoid mistakes and delays of human decision-making whendecision-makers are confronted with complex decision problems, meanwhile, itprovides relevant assistant decision-making information in detail fordecision-makers in decision problems so as to enable decision-makers to make ascientific decision-making.