Predicting Multi-KPI Warehouse Dynamics: A Simulation-Based Evaluation of Fuzzy-Integrated Neural Architectures

Warehouse order fulfillment is governed by complex, nonlinear interactions among order arrivals, picking workload, congestion, replenishment, labor, equipment, and dispatch capacity. This study proposes a simulation-based predictive framework that compares a global multilayer perceptron (MLP) with a fuzzy-integrated modular architecture comprising three regime-specific local neural networks. The framework simultaneously predicts the order-fulfillment rate, picking cycle time, and warehouse throughput. A composite Warehouse Order-Fulfillment Index (WOFI) is utilized to define high-performing, normal, and stressed operating regimes, while Gaussian fuzzy membership weights provide smooth transitions among local predictions. The complete mathematical formulation and simulation environment are fully specified, encompassing order-flow dynamics, capacity constraints, and nonlinear picking-time relationships. Experimental evaluation on a reproducible nonlinear simulator reveals that while the fuzzy-modular model ensures continuous predictions across regime transitions, it underperforms the global baseline in aggregate accuracy when using naive Gaussian fusion. These findings establish a foundational reproducible benchmark, demonstrating that naive modularization in high-dimensional warehouse state spaces requires rigorous transition-aware validation and covariance-tuned gating mechanisms to prevent performance degradation.

@article{Frikha2026,
  title     = {Predicting Multi-KPI Warehouse Dynamics: A Simulation-Based Evaluation of Fuzzy-Integrated Neural Architectures},
  journal   = {International Journal of Advanced Computer Science and Applications},
  volume    = {17},
  number    = {9},
  year      = {2026},
  publisher = {The Science and Information Organization},
  author    = {Mohamed Amine Frikha},
  doi       = {10.14569/IJACSA.2026.0170934},
  url       = {https://doi.org/10.14569/IJACSA.2026.0170934}
}