| 학회 | 한국화학공학회 |
| 학술대회 | 2006년 봄 (04/20 ~ 04/21, 대구 인터불고 호텔) |
| 권호 | 12권 1호, p.88 |
| 발표분야 | 공정시스템 |
| 제목 | Scheduling with virtual demand profiles for chemical plant |
| 초록 | Daily demand in the scheduling horizon is one of the most important input data for production scheduling. In real plants, however, collecting all demand information for the whole time horizon and forecasting the daily demand by traditional forecasting methods are very difficult. After all, the daily demand fluctuates according to market prices and customers, and it is hard to get a general trend of the daily demand. In this study, the concept of the virtual demand profile (VDP) is introduced and VDP is used for production scheduling. VDP is the demand profile built by probabilistic distribution and random values through analyzing historical demand data. In order to make the VDP better represent the real demand, several or hundreds of the VDPs are used for the scheduling problem in this approach. The VDPs are applied to various scheduling cases to show the advantages of using VDPs for production scheduling. Acknowledgement: This work is supported by Center for Ultramicrochemical Process Systems sponsored by KOSEF. |
| 저자 | 강민구1, 박선원1, Jay H. Lee2, Matthew J. Realff3 |
| 소속 | 1한국과학기술원 생명화학공학과, 2Georgia Institute of Technology, 3School of Chemical and Biomolecular Engineering |
| 키워드 | scheduling; PVC; virtual demand profile; stochastic demand forecasting |
| VOD | VOD 보기 |
| 원문파일 | 초록 보기 |
