We know that, a particular variable may have some inherent sub-domains out of which few of them may be intersecting or non intersecting. If such variable exists in a nonlinear objective function then it is hardly possible to optimize this function. Thus we need to create a unified domain by utilizing such subdomains. Then the function under study can be easily solved. However, we knew that fuzzy system can give an approximated solution of a function which is associated to the variables of impreciseness( non-random uncertainty). Thus using fuzzy system we may create almost nearly an unified domain.
I think the study area on fuzzy approximate reasoning can answer this question.
You may able to model your problem using disjunctive programming (https://en.wikipedia.org/wiki/Extended_Mathematical_Programming#Disjunctive_programming). AMPL/GAMS should support that.
Solution procedures depends on the complexity and dimension. One can use heuristic algorithms to solve the problem. Look into procedure...
10.13140/RG.2.2.12252.31361
Dear Bhattacharjee, I am asking for a numerical example based published work in the related field. Many thanks for your initiatives.
All the best.
- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Programming Languages