Applying ANFIS to Evaluate Demand of Collaborative Platform in Global Logistics

• Main Authors: LIN, GUO HUEI, 林國暉
• Other Authors: WANG, JIH CHANG
• Format: Others
• Language: zh-TW
• Published: 2005
• Online Access: http://ndltd.ncl.edu.tw/handle/99308390147433337913

碩士 === 長庚大學 === 資訊管理研究所 === 93 === The purpose of this study is to use “Adaptive Network-based Fuzzy Inference System (ANFIS)” to build the relation models based on the data sifted from the statistical data gathered by “Ministry of Finance” and “Directorate-General of Budget, Accounting and Statistics, Executive Yuan” of Taiwan, and then use these models to predict the estimation of future demands. We sift out the variables with closer relations from GDP, the foreign trade, and the amount of Customs’ manifests in Taiwan by linear regression. Then, the fuzzy inference system of ANFIS can help us building relation models which are able to predict the foreign trade with GDP or predict the amount of Customs’ manifests with import and export trades. At last, use R-square to check the models are usable or not. When all models are finished, the relation models can be used to estimate the ranges of demands with the optimistic GDP and the pessimistic GDP. Because the standard of the amount of Customs’ manifests was changed in 2002, we translate the calculated standard value, the real value divided by predicted value, into the final result. After finishing all steps mentioned above, we predict the future amount of Customs’ manifests of Taiwan until 2011. The optimist demand is 14186360 and the pessimistic demand is 14123144.