A note on connection between PNS and set covering problems
Process network synthesis (PNS) has enormous practical impact; however, its mixed integer programming model is tedious to solve because it usually involves a large number of binary variables. Using a combinatorial approach, a structural model of PNS can be given, and a branch-and-bound technique can...
Saved in:
| Main Authors: | |
|---|---|
| Format: | Article |
| Published: |
1996
|
| Series: | Acta cybernetica
12 No. 3 |
| Kulcsszavak: | Számítástechnika, Kibernetika |
| Subjects: | |
| Online Access: | http://acta.bibl.u-szeged.hu/12563 |
| Summary: | Process network synthesis (PNS) has enormous practical impact; however, its mixed integer programming model is tedious to solve because it usually involves a large number of binary variables. Using a combinatorial approach, a structural model of PNS can be given, and a branch-and-bound technique can be applied for searching an optimal solution. In some realistic examples of PNS, this method is efficient. Nevertheless, efficient methods are unavailable for solving these models generally. In this note, we describe a special class of PNS-problems as set-covering or set-partitioning problems. These problems are well-known to be NP-complete, thus, a PNS-problem is NP-hard. |
|---|---|
| Physical Description: | 309-312 |
| ISSN: | 0324-721X |