| Project/Area Number |
15510123
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Social systems engineering/Safety system
|
|---|
| Research Institution | Shizuoka University |
|---|
Principal Investigator |
SEKITANI Kazuyuki Shizuoka Univ., Fac.of Engineering, Assoc.Professor, 工学部, 助教授 (60256667)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
ANDO Kazutoshi Shizuoka Univ., Fac.of Engineering, Assoc.Professor, 工学部, 助教授 (00312819)
|
|---|
| Project Period (FY) |
2003 – 2005
|
| Project Status |
Completed (Fiscal Year 2005)
|
| Budget Amount *help |
¥2,000,000 (Direct Cost: ¥2,000,000)
Fiscal Year 2005: ¥500,000 (Direct Cost: ¥500,000)
Fiscal Year 2004: ¥600,000 (Direct Cost: ¥600,000)
Fiscal Year 2003: ¥900,000 (Direct Cost: ¥900,000)
|
|---|
| Keywords | Mathematical programming / Optimization model / AHP / ANP / DEA / Operations Research / Engenvalue problem / 相互評価 / サプライチェーン / 意思決定 / 合意形成 / 相互評価システム / RTS |
|---|
| Research Abstract |
This purpose of this study is to develop optimization modeling approaches to mutual evaluation systems that are Analytical Hierarchy Process (AHP), Analytical Network Process(ANP) and Data Envelopment Analysis(DEA). This study is carded out by theoretical approaches on the mathematical programming and empirical studies for managerial decision making problems.
We build a class of optimization models based on the eigenvector method and a sensitivity analysis for AHP/ANP. This approach is examined for personnel evaluation and a strategic decision making problem of the Supply Chain Management. We round that the conventional method provides a strange output that is not affected by a part of input data. A sufficient condition for the non response output is given. It is well known that an inner dependence among input data on the personnel evaluation is occurred. By incorporating Bayes theorem into the eigenvector method, we develop an optimization model that is effective to remove the inner dependence. Furthermore, it is free from a contradiction on a group consensus that is occurred in the conventional method.
We develop a new DEA model in time-series data and a single step measurement of Returns to Scale(RTS). The new DEA model for time-series analysis is well applied to measure relative efficiency among electric power companies. The single step RTS measurement has an advantage over the conventional multiphase measurement of RTS because the multiphase analysis occurs bias errors among the previous phase and the current one
|
