A factorisation-based recommendation model for customised products configuration design

Product configuration design combines the existing configurable components of enterprises to quickly form a new product that meets customised requirements. However, the representation, maintenance, and update of the configuration knowledge, and the mismatch problem restrict the rationality and efficiency of existing configuration design methods. In this paper, a recommendation model is developed for customised product configuration design, which takes personalised customer requirements and product component information as input and outputs a ranked list of component instances. It consists of two sub-models: a retrieval sub-model and a ranking sub-model. The retrieval sub-model selects a set of component instance candidates from all possible candidates, and then the ranking sub-model ranks them and selects the best possible candidate. To boost the ranking sub-model performance, we propose a novel interacting network, DualAdap, to extract meaningful low-order, high-order, and adaptive-order cross features. Based on learned cross features, the ranking sub-model computes the adoption scores of all candidates and then selects the best possible candidate according to adoption scores. The configuration design of the elevator traction machine is taken as a case study. Results verify that our recommendation model can identify similar component instances and then accurately pick out the best possible candidate that meets customer requirements.