ABSTRACT
Quantitative modelling, or mathematical modelling, has a history which goes back much further than the computer, though the computer has allowed models of greater complexity to be built. The behaviour of many simple systems can be predicted with great accuracy using just a few basic physical principles, given a knowledge of the initial conditions. Thus, Newton’s Laws can be used to make predictions about mechanical systems, if initial values of variables such as velocity, distance and force are known. In all quantitative models, the initial conditions are specified by giving values to independent variables; the model uses algebraic relationships to calculate the values of the dependent variables. So, the behaviour of the model depends on both the values of the independent variables and on the nature of the relationships between the variables. In complex systems, it may be difficult to identify and measure relevant variables, or to specify relationships between them. Nevertheless, quantitative modelling has taken on increasing importance in the social sciences. One ambitious and well-publicized attempt was described in The Limits to Growth (Meadows et al., 1973), in which a computer model of the social and economic system of the world was constructed and used to make long-term predictions about its possible behaviour.
