Travel time prediction in transport and logistics

Abstract

Purpose Many transport and logistics companies nowadays use raw vehicle GPS data for travel time prediction. However, they face difficult challenges in terms of the costs of information storage, as well as the quality of the prediction. This paper aims to systematically investigate various meta-data (features) that require significantly less storage space but provide sufficient information for high-quality travel time predictions. Design/methodology/approach The paper systematically studied the combinatorial effects of features and different model fitting strategies with two popular decision tree ensemble methods for travel time prediction, namely, random forests and gradient boosting regression trees. First, the investigation was conducted using pseudo travel time data that were generated using a pseudo travel time sampling algorithm, which allows generating travel time data using different noise processes so that the prediction performance under different travel conditions and noise characteristics can be studied systematically. The results and findings were then further compared and evaluated through a real-life case. Findings The paper provides empirical insights and guidelines about how raw GPS data can be reduced into a small-sized feature vector for the purposes of vehicle travel time prediction. It suggests that, add travel time observations from the previous departure time intervals are beneficial to the prediction, particularly when there is no other types of real-time information (e.g. traffic flow, speed) are available. It was also found that modular model fitting does not improve the quality of the prediction in all experimental settings used in this paper. Research limitations/implications The findings are primarily based on empirical studies on limited real-life data instances, and the results may lack generalisabilities. Therefore, the researchers are encouraged to test them further in more real-life data instances. Practical implications The paper includes implications and guidelines for the development of efficient GPS data storage and high-quality travel time prediction under different types of travel conditions. Originality/value This paper systematically studies the combinatorial feature effects for tree-ensemble-based travel time prediction approaches.

Travel Time Prediction in a Multimodal Freight Transport Relation Using Machine Learning Algorithms

“…This process minimizes the variance of the prediction results [41] (pp. [3][4]. In this study, regularization parameter, random_state, is also considered, which initiates the random number generator to randomize characteristics in trees [42] (p. 619).…”

“…can be scheduled accordingly. Consequently, manufacturers and logistic service providers can enhance their efficiency, optimize their processes, and increase planning accuracy [1,3,4].…”

Minimizing Traffic Congestion Through Continuous-Time Route Reservations With Travel Time Predictions

“…Notably, when there is large inconsistency between the observed travel times and travel time predictions then the performance achieved by DTA diminishes [19]. Clearly, the aforementioned issue can impact other state-dependent schemes and thus a large volume of work has focused on achieving accuracy in travel time predictions; using analytical or statistical methods [20]. Both methods, can be employed to predict travel times in a deterministic or stochastic manner [21], [13], always using data obtained through various traffic surveillance sensors (e.g., loop detectors, mobile detectors, radars and cameras [22]).…”

“…Statistical methods, are mainly data driven techniques based of current travel time observations [20], [23], [24]. It should be noted here that all the aforementioned methods only use the travel time measurements of the corresponding road segments without considering factors affecting travel times such as driver interactions and delays observed at intersections [20]. Along these lines, the work presented in [25] indicates that route travel times are affected by the segment's traversal time and the delays observed at the intersections (expressed as travel time penalties).…”

Macroeconomic Effects on Development of Sparsely Populated Areas in Public Transport

“…Average population in the working micro region 40164 Average number of economically active people in the working micro region 17320 Average number of occupied jobs in the working micro region 17854 Average number of municipalities in the working micro region 29 Average population in the centres of micro regions 23201 Average population in the centres of micro regions 5121333 Average number of economically active people in the centres of micro regions 10509 Total number of economically active people in the centres of micro regions 2995487 Average number of occupied jobs in the centres of micro regions 14109 Total number of occupied jobs in the centres of micro regions 2785456Source: authors a share of employers in the number of economically active people in the municipality and a share of the number of self-employed persons in the number of economically active people), unemployment rate, tax yield per capita and demographic indicators and a share of people commuting to work outside the municipality in the population of the municipality).The supplementary explanatory factors which we used in factor analyses included the population of municipalities and the number of occupied jobs in micro regional and interregional centres[25][26][27][28][29].…”