Aggregate planning in decentralized organizations

This paper deals with aggregate planning problems in decentralized organizations, i.e. organizations where the suborganizations have a fair amount of autonomy in control because information load, uncertainty and fast changing circumstances limit the possibilities of centralized control. The functions of aggregate planning, its place in the overall control structure of the company and the control variables involved are discussed and subsequently the additional complications of a decentralized setting. As a typical example of aggregate planning problems in a network of semi-autonomous organizations the so-called ‘internal business cycle’ is discussed: due to this phenomenon the network is not able to follow smoothly its final demand but moves from shortages in one period to surplusses in another. The causes, various positive feedback mechanisms through variations in the buffers between the units, are analyzed and illustrated with a simple simulation model. The paper concludes with a discussion on the use of aggregate planning to improve the control of the network and to dampen this internal business cycle.