Distribution Models for the Journey to Work

Abstract
The paper describes a model, based on random utility theory, for simulating the choices of home and workplace made by individual workers in a region. Each worker chooses the home and job which maximises his utility (comprising rent, salary, cost of the journey to work, and a random personal element); rents and salaries are adjusted iteratively until the distribution of workers’ choices of home and workplace matches the existing distribution of homes and workplaces. The model is applied first to several idealised urban geometries; it is then applied to 1971 Census data for the Manchester area, where the model reproduces satisfactorily the observed pattern of choices. Some of the effects of changed travel costs and of alternative policies for redistributing homes or jobs are then simulated.