The influence of aggregation level and category construction on estimation quality for freight trip generation models

Abstract : This paper analyzes the impacts of aggregation level and category construction on the relevance and quality of freight trip generation (FTG) models. More precisely, constant generations and functional form models are compared, as well as activity and activity-workforce categories. The paper proposes a method to compare constant generation and functional form models on different category classifications based on MAPE estimations. Functional forms are assessed via linear regression and compared using Pearson coefficient. Results show that the aggregation level has not always a positive impact on the model’s accuracy and the choice of suitable functional form leads to more accurate models.
Type de document :
Article dans une revue
Transportation Research Part E: Logistics and Transportation Review, Elsevier, 2019, 121, pp.134-148. 〈10.1016/j.tre.2018.07.007〉
Liste complète des métadonnées

https://hal-emse.ccsd.cnrs.fr/emse-01860647
Contributeur : Florent Breuil <>
Soumis le : jeudi 23 août 2018 - 15:24:58
Dernière modification le : jeudi 7 février 2019 - 17:11:29

Identifiants

Citation

Jesus Gonzalez-Feliu, Iván Sánchez-Díaz. The influence of aggregation level and category construction on estimation quality for freight trip generation models. Transportation Research Part E: Logistics and Transportation Review, Elsevier, 2019, 121, pp.134-148. 〈10.1016/j.tre.2018.07.007〉. 〈emse-01860647〉

Partager

Métriques

Consultations de la notice

97