AI to Detect Firms’ Competitive Performance

Artificial Neural Networks (ANNs) are compelling tools to support managers in making decisions and scholars studying relevant phenomena like reshaping business and organizational processes, extracting relationships between value drivers and economic value added, human resources management, workforce substitution, financial risk prediction, and distress analysis. However, there is still room for further studies on ANNs to identify the critical determinants of companies’ performance and to simulate the effects of a change in both external (i.e., climate change, events, infrastructures) and internal (prices, distribution channels, segments, product features) drivers on firms' results. This research wants to bridge this gap by presenting the outcomes of 2 related studies. The first focuses on identifying key drivers of competitive performance, and the second focuses on dynamic simulations. We tackle both studies with a specific unsupervised ANN, Auto Contractive Map, which outperforms other methods in detecting internal and external drivers of firms' competitive results. Auto Contractive Map (CM) application brings to light hidden fundamental connections between variables, showing which variables are directly connected and which ones are mediated by other variables. Moreover, dynamic simulations demonstrate Auto Contractive Map’s effectiveness in forecasting the impact on companies' results of a change in specific internal or external determinant of competitive performance. The selected area of investigation is the hospitality industry, where external performance drivers are particularly relevant due to hotels' strong dependency on the local ecosystem.