Papers by Tomasz Romanowicz
Technological Forecasting and Social Change, 1988
The conversational computer model of the dynamics of car populations, presented in Part I of this... more The conversational computer model of the dynamics of car populations, presented in Part I of this paper, is shown against the background of actual data. First, contents of the model are outlined, and its basic assumptions are illustrated with the data available. Then, model runs are cited to provide a basis for comparative analysis of actual and theoretical behavior and potential causes of eventual divergences. 1. Introduction-Outline of the Model Part I of this paper contains a description of a conversational computer model developed to imitate the dynamic behavior of numbers of passenger cars according to their "types" and "ages." The structure of the model is derived from mathematical formulae describing originally the dynamics of biological populations according to genetic adaptive fit and age cohorts. In its complete formulation, i.e., accounting for both nontrivial age and type dimensions, such a dynamic model cannot be solved analytically, and that is why it was developed as a conversational simulation tool. Thus, the model can easily be run on any PC or even sub-PC hardware, updated on-line, rerun, etc. The fundamental assumptions behind the model are as follows: Cars are classified along their ages and types into age-type cohorts, within which they are not distinguished, and the model simulates the changes of car numbers within these cohorts. The main driving force of the "car population" dynamics is the observation that the numbers of new registrations approximates well the number of cars disappearing from the population (i.e., scrapped). The dynamics, governed primarily by the assumption above, is therefore "shaped' by the "mortality" characteristics of particular car types; in fact, car types in the model differ just in that respect. On top of the two previous assumptions, various hypotheses can be tested relating variables both endogenous and exogenous to the model. In the concrete implementation of the model, as shown in Part I, these assumptions take the following practical forms:
Lecture Notes in Control and Information Sciences
Without Abstract
Uploads
Papers by Tomasz Romanowicz