Inventions are essentially knowledge search (KS) processes in which prior knowledge is sought, acquired and recombined. However, there is inadequate quantitative modelling of KS rules and costs in invention processes for expediting technological breakthroughs. This study, therefore, proposed a dynamic model to simulate KS processes for real inventions with network-theoretic and heuristic approaches. We constructed the prior knowledge network (PKN) of a focal patent as the search space, formulated heuristic search rules and designed estimates for KS costs. Simulations in the field of photolithographic technology showed that search costs differed significantly with different heuristics. Heuristics based on knowledge familiarity and importance significantly outperformed other rules in terms of KS costs and were less affected by the size and density of PKNs. Interestingly, there was no significant correlation between the mean and variance of search costs and patent value, indicating that high-value patents are not particularly difficult to obtain. This study broadens the scope of current information-seeking models by considering knowledge-creation behaviour and offers practical guidance for increasing R&D efficiency.
Ren et al. (2025) studied this question.