| Abstrakt: |
Amidst the rapid urbanization and the consequent surge in urban population on a global scale, the significance of efficient transportation systems has never been more pronounced. This is particularly true in critical sectors like humanitarian aid, healthcare, and pharmaceutical logistics, which face unique challenges and costs that deviate from the usual logistical norms. Strikingly, in Morocco, there's a notable absence of comprehensive studies on pharmaceutical transportation, particularly concerning the associated costs and delivery conditions. This glaring gap in research underscores the pressing need for the development of a tailored model that squarely addresses these issues. Pharmaceutical transportation presents a multifaceted landscape characterized by high-dimensional regression or classification challenges. It's further complicated by the intricacies of variable selection, especially when dealing with interrelated predictors. In this context, the Random Forests algorithm emerges as an appealing solution for both classification and regression tasks. It has demonstrated robust predictive performance and the capacity for variable selection through importance measures. In this comprehensive manuscript, we propose an innovative cost prediction model specifically tailored for pharmaceutical transport within Morocco. To set the stage for this model, we embark on a theoretical exploration of the significance of permutation importance within the context of additive regression models. This endeavor offers insights into how the correlation between predictors influences the importance of permutations. Building on this theoretical foundation, we proceed to establish our predictive cost scheme. Our model exhibits a commendable predictive performance, surpassing an accuracy threshold of 75%. This achievement underscores the robustness of the Random Forests algorithm in capturing the complexities of transportation. This multifaceted approach to cost prediction within the realm of pharmaceutical transportation in Morocco stands to provide valuable insights and practical solutions for this critical sector. [ABSTRACT FROM AUTHOR] |
