Use this URL to cite or link to this record in EThOS:
Title: Techno-economic forecasting for packaging materials
Author: Thomas, M. C.
Awarding Body: University of Wales Swansea
Current Institution: Swansea University
Date of Award: 2001
Availability of Full Text:
Access from EThOS:
Forecasting packaging material demand is crucial to effective future planning by capital intensive material manufacturers. Yet several considerations introduce great uncertainty over the future packaging mix. Foremost is a highly heterogeneous and dynamic end-use marketplace subject to multitudinous technological, economic, consumer and legislative change-forces. These act at all levels upon a diverse and complex supply chain that suffers data paucity and, hence, opacity of cause and effect. A wide range of future-oriented decision technologies was examined to meet these challenges. None promise competitive advantage over commissioned forecasts of aggregate demand. At the sector level, the petfood market is relatively homogeneous and simple, but nonetheless significant. Neural network analysis of its causal relationships led to rich results and a simple, workable causal-forecasting model. Data paucity inspired three key development paths. First, a weakness in current implementations of genetic algorithm model input selection was exposed - result variability due to training data set division. Novel software invoked genetic algorithm input selection over exhaustive permutations of training cases to generate a result distribution, thereby partially automating model specification for wider application. Second, the neural networks were implemented in a scenario-planning spreadsheet, to isolate the more certain and less certain factors in scenario forecasting. Third, several unprecedented factors change past relationships and can undermine even the most accurately specific model. Accordingly, a Delphi survey was conducted to develop scenarios of the potential impact of remote retailing upon packaging demand. Consequently, although the five-year outlook for tinplate petfood packaging is open to interpretation, the most likely scenario is stable demand. Petfood and human food cans exhibit clear strengths in the remote-retailing scenario, but high uncertainty is envisaged for the remaining packaging applications. Such unprecedented forces should be continually monitored, and marketing activities should emphasise the strengths of tinplate in the scenarios thereby envisaged.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (D.Eng.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available