A Dynamic and Adaptive Decision Support System for Managerial Time Allocation Under Environmental Uncertainty

Abstract

Introduction: Managerial time allocation is a critical mechanism through which managers influence organizational performance and adaptability. Existing decision support approaches typically treat time allocation as a static, one-shot optimization problem, despite the fact that managerial priorities, constraints, and environmental conditions evolve continuously. In uncertain and dynamic environments, static recommendations risk becoming misaligned with organizational needs, reducing their practical usefulness.

Objectives: The objective of this study is to develop and evaluate a dynamic and adaptive decision support framework for managerial time allocation that explicitly accounts for environmental uncertainty and temporal change. The study aims to assess whether adaptive time-allocation policies outperform static and purely reactive approaches in terms of cumulative value, robustness, and behavioral stability.

Methods: Managerial activities are modeled as a time-dependent portfolio, with time allocations revised periodically using a rolling-horizon decision process. Managerial preferences are represented through interval-valued parameters to preserve flexibility, while environmental uncertainty is captured via stochastic states and scenario-based evaluation. The resulting optimization problem is solved using a robustness-aware rolling-horizon evolutionary algorithm. Performance is evaluated through simulation under multiple uncertainty regimes, including low volatility, high volatility, shock-driven, and regime-switching environments, and compared against static and non-robust rolling-horizon benchmarks.

Results: The results show that dynamic adaptation provides limited benefits in stable environments but yields substantial performance improvements under volatility, shocks, and regime changes. The proposed adaptive robust model consistently achieves higher cumulative value and lower performance variability than benchmark approaches. Dynamic analysis further reveals that the model selectively adapts to meaningful environmental changes while avoiding excessive oscillations in time allocations.

Conclusions: The findings demonstrate that managerial time allocation should be supported by adaptive, uncertainty-aware decision support systems rather than static schedules or purely reactive adjustments. By formalizing time allocation as a dynamic portfolio problem, this work contributes to the decision support and management science literature and provides a foundation for intelligent systems that evolve alongside managerial environments.