Multi-objective optimization of reliability-redundancy allocation problem for multi-type production systems considering redundancy strategies

A novel boundary swarm optimization method for reliability redundancy allocation problems

“…If the number of local nondominated solutions (Nlns) is larger than Nrep, these local nondominated solutions with smaller crowding distances are discarded until the number of Nlns does not exceed that of Nrep. The concepts of the repository and the crowding distance are adopted from MOPSO [44] and NSGA-II [10, [14][15][16], respectively, and are combined to utilize the advantages of both in the MOSSO. To fix this problem in the MOPSO, Xi is always added to the repository, and the old Pi is replaced with the new Pi if Xi and the old Pi are not dominated by each other in the MOSSO for i = 1, 2, …, Nsol.…”

“…Because many distinct types of systems are employed in various fields, there are different categories of RRAPs with different redundancy strategies, i.e., active strategies [5,6,8,9,11,13,14,[17][18][19], standby strategies [7,10,15], and mixed strategies [12,16]. The differences between these strategies are discussed briefly below.…”

“…To solve this well-known NP-hard problem, various artificial-intelligence techniques have been widely studied in the past decades for numerous systems with diverse considerations [5][6][7][8][9][10][11][12][13][14][15][16][17][18][19]. Examples of such techniques include stochastic fractal search (SFS) [17], genetic algorithm (GA) [10], a hybrid algorithm of GA and PSO [13], particle swarm optimization (PSO) [9], stochastic perturbation PSO [12], the artificial bee colony algorithm [6], cuckoo search [18], particle-based simplified swarm optimization Maximize Rs(n, r) (1) Subject to gv(n, r)≤ Vub (2) gc(n, r)≤ Cub (3) gw(n, r)≤ Wub (4) nlb ≤ n=(n1, n2, .…”

“…10,[14][15][16]. Let Xt,1, Xt,2, …, Xt,Nsol be solutions and Xt,Nsol+1, Xt,Nsol+2, be offspring (solutions) generated by parents randomly selected from {Xt,1, Xt,2, …, Xt,Nsol} in the generation t. The major difference between the GA and NSGA-II is the method of selecting solutions Xt+1,1, Xt+1,2, …, Xt+1,Nsol from Ω = {Xt,1, Xt,2, …, Xt,2Nsol} for generation (t+1).…”

Fuzzy reliability‐redundancy allocation problem of the overspeed protection system

“…A dynamic self‐adaptive multiobjective particle swarm optimization method was developed by Khalili‐Damghani et al 13 to solve binary‐state multiobjective reliability‐redundancy allocation problems, where a penalty function and modification strategies were used to handle the constraints in the problem. Wang et al 14 modeled a multitype production system as fuzzy multiobjective reliability‐redundancy allocation problem, subject to resource constraints. The structure of the production system was modeled by introducing a binary matrix, and a multiobjective evolutionary algorithm was suggested to solve the optimization problem.…”