Indian Wedding System Optimization (IWSO): A Novel Socially Inspired Metaheuristic with Operational Design and Analysis

Technical Summary: Indian Wedding System Optimization (IWSO)

Problem Statement
Metaheuristic optimization algorithms are essential for solving complex, nonlinear, and high-dimensional problems in engineering and computational intelligence. However, classical approaches such as Genetic Algorithms (GA), Particle Swarm Optimization (PSO), and Differential Evolution (DE) frequently encounter persistent challenges, including premature convergence, loss of population diversity, and limited adaptability to dynamic environments. These limitations often stem from fixed search parameters and rigid population dynamics, which constrain performance in large-scale or time-varying optimization tasks. The paper addresses the need for an adaptive, socially inspired framework capable of maintaining diversity and balancing exploration with exploitation more effectively than existing nature- or physics-based analogies.

Methodology
The authors propose the Indian Wedding System Optimization (IWSO), a novel population-based metaheuristic inspired by the socio-cultural dynamics of traditional Indian weddings. The algorithm models the matchmaking process—characterized by collaboration among families, candidates, and matchmakers—as a guided, selective search framework.

The operational design of IWSO follows a seven-step iterative cycle:

1. Initialization: A population of candidate solutions is generated randomly within defined bounds, analogous to families considering multiple potential matches.
2. Fitness Evaluation: Candidates are evaluated using a weighted multi-objective fitness function that aggregates multiple criteria (e.g., compatibility, financial stability), mirroring the multi-factor decision-making in matchmaking.
3. Best Candidate Identification: The optimal solution (the "best match") is identified to serve as a benchmark.
4. Matchmaker-Guided Influence: An adaptive matchmaker factor ($M$) controls the update of candidate positions. This factor decreases over iterations, shifting the algorithm from broad stochastic exploration to intensified local exploitation. Elite solutions guide weaker candidates toward promising regions without requiring external control parameters.
5. Adaptive Elimination and Reinitialization: To prevent stagnation, candidates that are too similar to the best match but not optimal are eliminated. They are replaced by new candidates generated through a combination of the global best solution, the population mean, and controlled stochastic perturbations. This mechanism maintains diversity and prevents premature convergence.
6. Selection: The best solution is updated based on fitness comparisons.
7. Termination: The process repeats until the maximum iteration count is reached or a convergence criterion is met.

Theoretical analysis establishes that IWSO exhibits linear time complexity of $O(T_{max}n(D + K))$ and space complexity of $O(nD)$, where $n$ is population size, $D$ is dimensionality, and $K$ is the number of objectives. Convergence analysis proves that the algorithm converges almost surely to the global minimum under specific conditions regarding the stochastic perturbations and elitist selection.

Key Contributions
The paper outlines four primary contributions:

• Novel Metaphor-Driven Framework: The introduction of IWSO, which maps socio-cultural components of Indian weddings (matchmakers, suitors, elimination rituals) to algorithmic operations for search, selection, and adaptation.
• Matchmaker-Guided Influence: A strategy where elite solutions direct the evolution of weaker candidates, enhancing convergence and search space exploration without reliance on external control parameters.
• Adaptive Elimination and Reinitialization: A mechanism that dynamically replaces underperforming individuals to maintain population diversity and prevent stagnation, addressing a common failure mode in classical metaheuristics.
• Comprehensive Evaluation: A rigorous theoretical and empirical analysis demonstrating IWSO's computational efficiency, scalability, and superiority over state-of-the-art (SoTA) swarm intelligence algorithms.

Experimental Results
The performance of IWSO was evaluated on 23 benchmark test functions (covering unimodal, multimodal, separable, and non-separable categories) and compared against ten established algorithms: GA, ABC, PSO, ACO, DE, WOA, HS, SA, BHO, and CS.

• Convergence and Stability: IWSO demonstrated superior convergence speed and solution quality across high-dimensional and multimodal landscapes. In statistical comparisons over 30 independent runs, IWSO achieved lower standard deviations on 18 out of 23 functions, indicating higher stability and reliability.
• Efficiency: The algorithm consistently outperformed competitors in terms of runtime (RT), often requiring significantly less computational time to reach optimal or near-optimal solutions.
• Sensitivity Analysis: The study analyzed the impact of control parameters, specifically the Matchmaker Factor ($M$) and iteration count ($T_{max}$). Results indicated that IWSO is robust across varying parameter settings, with the adaptive mechanisms effectively balancing exploration and exploitation regardless of problem complexity.
• Comparative Performance: In 3D surface plots analyzing fitness values against population size and iteration counts, IWSO consistently guided the search toward lower prediction errors compared to GA and DE, confirming its adaptability and precision.

Significance and Claims
The paper claims that IWSO represents a significant advancement by bridging human-inspired coordination with computational intelligence. Unlike conventional nature-inspired algorithms that rely solely on physical or biological analogies, IWSO introduces a socially grounded optimization paradigm. The authors assert that by embedding social hierarchy, cooperative learning, and adaptive feedback into its core design, IWSO achieves sustained population diversity and robust convergence even in complex, constrained, and high-dimensional search spaces.

The work positions IWSO as a robust alternative to conventional metaheuristics, particularly for scenarios requiring fast convergence, resilience to stagnation, and adaptability in dynamic environments. The authors suggest that the algorithm's ability to handle multi-objective and high-dimensional problems makes it suitable for real-world large-scale optimization tasks, though they note that future research will focus on hybrid extensions and applications in distributed and time-varying settings such as cloud resource orchestration and industrial scheduling.