Comparative efficacies of decision strategies and the effects of learning in dynamic environments: a computer simulation approach

Abstract

Models of aggregation in management science and economics are not consistent with micro-empirical knowledge of individual decision making. This has occurred as a result of using heuristics that are derived from behavioral studies which focused on discrete incidents. This approach fails to recognize decision making as a continuous process and overlooks the importance of feedback. This study examines the performance of various decision strategies (heuristics) in dynamic environments through computer simulation. Within dynamic task environments, three classes of strategies are examined: (a) feedback oriented strategies, (b) non-feedback oriented strategies and; (c) a strategy that incorporates learning. The relative efficacies of these strategies are compared. The results show that feedback oriented strategies achieved a higher level of performance than non-feedback oriented strategies. And the strategy that incorporated learning outperformed all other strategies. A few anomalies exist and may require additional sampling. The implications of these findings for command decision making indicate that, feedback from prior military actions can play an important role in adapting existing systems to meets new military roles in changing environments.