Computational Models and Their Role in Problem Solving

Abstract

Computational models of systems—abstractions of how components interact to produce collective phenomena—assist with problem solving in various domains. Tasks include exploring systems qualitatively or quantitatively to generate insight, evaluating consequences of proposed interventions, or devising interventions aimed at achieving specified objectives. Inputs to a model can consist of parameters, data, or initial conditions that represent the knowledge and uncertainties surrounding the system and its behavior, whereas outputs indicate model predictions. Models can thus be employed to design experiments, validate theoretical descriptions, or support diagnoses.

Computational problem solving spans multiple disciplines, each with differing characteristics, team configurations, and development stages. In the natural sciences, models may aid in the development, analysis, or implementation of new theories; selection of variables for experimental investigation; inference of unknown parameters from empirical data; prediction of behavior under varying assumptions; or evaluation of impact of environmental change. In engineering and technology, models are used to evaluate alternative configurations or designs; assess functional or performance requirements; consider reliability or safety factors; or predict failure locations or mechanisms. Within the social sciences and humanities, computational tools address societal problems; examine historical contexts, parallels, or human behavior; study the efficacy of different policy options; or analyze artworks. In biomedicine and health, they evaluate drugs, treatments, or interventions; assist in diagnosis, triaging, or screening; interpret signals or images; or help resolve disputes in legal cases (Noranis Mohd Aris & Azuan Nazeer, 2013) ; Poesia & D. Goodman, 2022; Al-Fedaghi & Abdullah Alkhaldi, 2019).

Keywords: Computational models, problem solving, multiple disciplines, efficacy of different policy.