Multi-criteria probability collectives
Neha S. Patankar, Anand Jayant Kulkarni, Kang Tai, TRUPTI GHATE -, A.R. Parvate · International Journal of Bio-Inspired Computation · 2014
The nature-/bio-/socio-inspired optimisation techniques can efficiently handle unconstrained problems; however, their performance gets significantly affected when applied for solving constrained problems. This paper proposes a variation of the distributed optimisation multi-agent system (MAS) approach of probability collectives (PC) in collective intelligence domain referred to as multi-criteria probability collective (MCPC). In this approach, the constraints are efficiently handled by giving equal importance as the objective function. It is validated by solving a variety of constrained test problems including tension/compression spring design problem and pressure vessel design problem. The solution to these problems proves that the MCPC approach can be applied to a variety of complex practical/real world problems.