Computer Simulation for Contemporary Problems in Civil Engineering

Abstract

This paper is intended to demonstrate that contemporary computer simulations in civil engineering are different to conventional simulations of the past. In the majority of cases, conventional computer simulations have been based on numerical simulation models, particularly on finite element models solving ordinary and/or partial differential equations. Though such numerical simulations, besides theory and experiment, are rightly considered as one of the columns of knowledge gain, they do not fit the needs of the today’s “big challenges”. Big challenges in civil engineering are characterized by a complex and intricate nature. More crucial, such tasks are founded on (i) a collaborative network of actors, (ii) a real world system or an abstract model of it , representing static, dynamic as well as functional (data flow) behavior patterns, and (iii) an environment. The environment acts upon the system or its model, but in turn also reacts to the behavior of the system or its model, as well as to the events of the system’s user, due to feedback mechanisms. To this end, the corresponding computer simulations require a multi-level and distributed approach where the individual levels themselves possess different scal es in time and space. An aggravating factor is that the representation of each level is based upon different solution paradigms, not only numerical but also non-numerical ones, e.g. data-based, knowledge-based, event-driven and/or object-oriented paradigms. In the following, specific applications are given to elucidate and illustrate how this widespread category of simulation is becoming not only a new perspective in computational engineering but also the mandatory rationale for solving current, and future, sophisticated problems. Upon a general disquisition of the topic, exemplary research applications are scrutinized.