Research on Product Positioning Optimization Modeling Based on Improved MNL and Its Solution by ABC

The diversification and individuation of users' requirements put forward higher requirements for decision makers to positioning products quickly and accurately. Therefore, this paper proposes an improved Multinomial Logit Model (MNL) for positioning design in the product development phase. For the vagueness of the customer's qualitative evaluation of the product, a cloud model is used to convert the qualitative evaluation into quantitative user preference data, that is, the evaluation cloud of the product attributes. The utility value of each attribute level is calculated according to the expected value of the evaluation cloud, and then the user preference value of the product is obtained by weighted accumulation, where the weights of product attributes are optimized by the expected value and entropy value of the evaluation cloud and therefore the subjectivity and inaccuracy of the current method of attribute weight set by experience is avoided. Then, the MNL model is used to calculate the customer selection probability of the product, and the product positioning model is established with the aim of maximizing the profitability of the product. Finally, the model is solved by the Artificial Bee Colony Algorithm (ABC), and then the optimized combination of product attribute levels is obtained to determine the optimal product profile. At the end of the paper, the feasibility and rationality of the product positioning model and its solution method are verified by an example experiment.