Heuristic-based management (I): variation.

Purpose - The paper considers model-based management and, based on it, proposes a heuristic-based management. This paper aims to assert that heuristic-based management, for complex systems, a process of free variation, of pairs of models and actions called organisational strategies, maximizes the chances of improving the system's performance in open environments. Design/methodology/approach - A conception of complex systems are introduced and characterized as open and self-organising systems. Then, the proposal to heuristically use pairs of models and actions, called organisational strategies, to manage social systems based on evolutionary thought is supported. Subsequently, a computational experiment is proposed to show that, even in a simple framework, variation processes are required. Findings - The paper shows that two processes may be required to preserve self-organising systems. This finding indicates that variation and selection processes, related to evolutionary thought, are necessary for managers to deal with complex systems interacting with complex environments. Finally, it is shown that, even in simple computational environments, variation may be required. Research limitations/implications - The paper is the first part of an ongoing research agenda on the subject of heuristic-based management and only refers to variation processes. Originality/value - The paper links complex systems theories to evolutionary thought. It also relates principles of cybernetics to those of game theory. The proposal has been formalized based on these relations, and has been called heuristic-based management. Principles first developed in information theory, organisational cybernetics, and evolutionary thought are used so that a complex system can be effective when interacting with a complex environment.