Tacit knowledge-based expert model for decision support in injection mould design

Injection moulding is the most popular technique used by Sri Lankan small and medium-scale enterprises for producing plastic parts in large quantities. Once a strong industry that catered to over 50% of local demand for dies and moulds are now barely meeting one-tenth of the demand due to import competition. Moreover, the local industry lacks access to state-of-the-art design tools owing to funding limitations and mainly relies on experience of long-standing mould design experts. The brain drain following country's economic downturn has now severely affected growth or even existence of this important industry. In this context, the paper introduces a unique strategy to digitally transform the tacit knowledge of mould design experts, and thereby establish a data-driven decision-making process for mould design. In order to manage both explicit and tacit knowledge in injection mould design, a framework was developed, and corresponding databases were established. The proposed expert model uses a predefined case bank and a case-based filtering algorithm to identify matching data sets for a given new design from explicit and tacit databases. Suitability of parameters of the new design is determined using a decision-making algorithm, where higher weightage is assigned to tacit knowledge-based on data availability. The expert model was validated using a case study, and results of the expert model were found to be significantly agreeable to the output of industry-standard mould design software. The findings indicate potential of the proposed tacit knowledge-based expert model to positively impact mould industry affected by low resources and help reach Industry 4.0.