A hybrid machine learning-optimization framework for modeling supply chain competitive pricing problem under social network advertising

Designing a system to improve the performance of the supply chain by linking pricing and advertising will bring significant benefits to the supply chain components. Due to the expansion of customers' activities in social networks and society's greater interest in obtaining information from this space, companies have changed their marketing methods. Organizations have turned to influencer marketing for advertising to take advantage of social networks' potential. Despite the recent growth of influencer marketing, the issue of optimal identification and appropriate spending based on the characteristics of influencers to convey a company message or advertisement has not been carefully studied in the academic literature, nor has it been addressed in practice. Therefore, in this paper, a hybrid framework of machine learning and optimization is developed to help organizations (i) conduct successful and optimal marketing campaigns by selecting the best influencers and (ii) examine the effectiveness of advertising on the pricing of products in the supply chain. The framework designed in this study is validated using data extracted from the Instagram platform. The results demonstrate that businesses should adopt novel approaches to get the maximum benefit from advertisements on social media, which depends on a more accurate selection of influential users. Contrary to previous works, our results demonstrate that choosing influencers based on their influence network will not necessarily bring the expected efficiency for businesses.

Automated summary

Designing a system that links pricing and advertising to improve supply chain performance can bring significant benefits. This study validates the importance of accurately selecting influencers for social media advertisements and suggests that choosing influencers based solely on their influence network may not yield the expected results.