Modelling Practice

This chapter deals with the practicalities of building, testing, deploying and maintaining models. It gives specific advice for each phase of the modelling cycle. To do this, a modelling framework is introduced which covers: problem and model definition; model conceptualization; model data requirements; model construction; model solution; model verification; model validation and finally model deployment and maintenance. Within the adopted methodology, each step is discussedthrough the consideration of key issues and questions relevant to the modelling activity. Practical advice, based on many years of experience is providing in directing the reader in their activities.Traps and pitfalls are discussed and strategies also given to improve model development towards “fit-for-purpose” models. The emphasis in this chapter is the adoption and exercise of a modelling methodology that has proven very successful in many model building activities. It is vital that good methodologies are adopted for both thoroughness and efficiency purposes. Asking good questions for each modelling stage can aid in getting to effective and efficient solutions in modelling practice. Modelling is very much a ‘goal oriented’ activity, under constraints of system insight, time, cost and human resources. The George Box dictum that “all models are wrong, some are useful” should be coupled with the parsimony principle to ensure optimal outcomes.