Prediction of Greener Last-Mile Delivery Adoption Intention in Telemedicine Supply Chain: A Machine Learning Approach

This study seeks to predict the intention to adopt greener last-mile delivery in telemedicine supply chain using machine learning-based approach. Data used in the study were acquired from 349 respondents in the Philippines, and examined using different machine learning techniques, namely, Gradient Booting, Random Forests, Support Vector Machines, K-Nearest Neighbor, XGBoost, and Decision Tree further validated using various performance metrics. Results demonstrated that more than 80% of machine learning models' performance accurately predict intention to adopt greener last-mile delivery in telemedicine. Moreover, RF, SVM, and XGB attained optimal prediction performance. Attitude towards greener delivery, perceived behavioral control, perceived usefulness, were the most significant factors influencing the intention to adopt greener last-mile delivery, followed by subjective norms and trust in technology. Interestingly, perceived ease of use ranks the lowest, indicating that intention to adopt greener last-mile delivery among individuals is mostly affected by attitude to support green logistics and their perceived benefits rather than the ease and trust of using this technology. Lastly, theoretical and practical implications, together with effects in telemedicine supply chain are presented at the end of the study.