A Design Model for Allocating Interrelated Land-Use Activities in Discrete Space

A fundamental problem confronting architects and land-use planners is allocating interrelated activities to sites in order to achieve a set of predetermined goals. In some planning situations, such as the design of new towns, all activities can be initially manipulated under very few constraints, and a large number of alternative plans are possible. The generation and selection of the ‘best’ plan is further complicated when we consider the interrelationships between the locational activities. This paper proposes a multistage planning approach capable of producing and selecting ‘best’ allocation plans. The structural features of the model are discussed and an optimization procedure is presented. The latter combines the exchange-elements allocation technique with the Newton-penalty-function optimization method. The model's proficiency as a planning tool is examined by use of data drawn from a real-world case study.