Butterfly Optimization Algorithm

Each butterfly’s perceived fragrance is linked to the value produced by the objective function. Stronger or weaker fragrance therefore changes how candidate solutions influence movement during the search. This connection converts engineering performance measurements into search guidance, allowing the algorithm to favor promising regions without calculating gradients of the design problem.

The sensory-modality parameter controls how strongly butterflies respond to perceived fragrance. Changing it alters the influence that objective-function performance has on movement, so parameter selection affects the balance between responsive guidance and broader exploration. Engineers must therefore treat sensory modality as a problem-specific setting rather than a universally fixed value.

Global and local search address different risks during optimization. Global movement helps examine widely separated regions of the solution space, while local movement focuses more closely on currently promising areas. Switching between them supports both exploration and refinement, which is important when an engineering problem contains competing design possibilities or complex objective-function behavior.

The method evaluates candidate solutions through their objective-function values instead of requiring derivatives that describe how performance changes with each variable. This makes it suitable for engineering formulations where gradient information is unavailable or impractical. Its population-based search can also represent continuous or discrete design variables, broadening the types of optimization problems it can address.

An application begins by representing feasible design alternatives as butterflies and evaluating each one with the chosen objective function. The search then updates candidates using fragrance-guided movement, alternates global and local search, and applies the selected sensory-modality setting. The resulting designs are compared through their objective values to identify the preferred engineering solution.

Engineers can apply the approach to structural design, parameter estimation, scheduling, energy management, and control-system tuning. Its usefulness depends on how the objective function represents the engineering goal and on suitable parameter settings, including the sensory-modality control. Consequently, performance is problem-specific, and a successful application requires matching the search configuration to the design task.