A machine learning approach for the operationalization of latent classes in a discrete shipment size choice model

Piendl, Raphael; Matteis, Tilman; Liedtke, Gernot

doi:10.1016/j.tre.2018.03.005

Cited by 11 publications

(5 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Second as stated by Train (2008), the estimation of a full set of parameters quickly reaches the limit of gradient descent approaches with respect to the computational complexity. Thus, freight transport seems to be a big challenge due to its heterogeneity and how it can be measured as already indicated by Piendl et al (2019). When forecasting, the demand also needs to be assigned to each segment (including the firm-to-firm commodity flows).…”

Section: Heterogeneity Of Logistics Requirementsmentioning

confidence: 99%

“…These methodologies are described shortly in the two following subsections. Piendl et al (2019Piendl et al ( , 2017 provide a more in-depth presentation.…”

Section: Modelling Frameworkmentioning

confidence: 99%

“…Finally, if there is no information about shipments attributes, the usability of the segments is limited. Nevertheless, if surveys contain information on the standardised classifications, one might apply a probabilistic assignment approach which accounts for the relation between standard classifications and latent segments, as proposed by Piendl et al (2019). If there is no information about correspondences at all, simple probabilistic assignments according to the estimated segment proportions are an option.…”

Section: Wider Implications Of the Transferability Analysismentioning

confidence: 99%

“…However, doing so would be missing the opportunity to leverage the information available in shipper surveys, and would provide little information regarding how freight data collection protocols should be improved in the future: this is one of these situations where improving the statistical fit of models and improving their practical usefulness are two distinct objectives. Finally, the approach comes with corollary qualities: in a companion article, Piendl et al (2019) showed how these segments could be matched with classical commodity flow data, with a machine learning approach, so as to simulate the impact of exogenous changes (information and communication technology in their article) on shipment sizes and transport operation. Now, before the introduction of handling and conditioning variables into freight database production and freight modelling state-of-the-art can be seriously considered, two questions need answers.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Building latent segments of goods to improve shipment size modelling: Confirmatory evidence from France

Piendl

Koning

Combes

et al. 2022

European Journal of Transport and Infrastructure Research

View full text Add to dashboard Cite

Freight transport demand models are generally based on administrative commodity type segmentation which are usually not tailored to behavioral freight transport demand modelling. Recent literature has explored new approaches to segment freight transport demand, notably based on latent class analysis, with promising results. In particular, empirical evidence from road freight transport modelling in Germany hints at the importance of conditioning and handling constraints as a sound basis for segmentation. However, this literature is currently sparse and based on small samples. Before it can be accepted that conditioning should be integrated in the state-of-the-art doctrine of freight data collection and model specification, more evidence is required. The objective of this article is to contribute to the issue. Using detailed data on shipments transported in France, a model of choice of shipment size with latent classes is estimated. The choice of shipment size is modelled as a process of total logistic cost minimization. Latent class analysis leverages the wide range of variables available in the dataset, to provide five categories of shipments which are both contrasted, internally homogenous, and directly usable to update freight collection protocols. The groups are: "‘standard temperature-controlled food products"’, "‘special transports"’, "‘bulk cargo"’, "‘miscellaneous standard cargo in bags"’, "‘palletised standard cargo"’. This segmentation is highly consistent with the empirical evidence from Germany and also leads to better estimates of shipment size choices than administrative segmentation. As a conclusion, the finding that conditioning and handling information is essential to understanding and modelling freight transport can be regarded as more robust.

show abstract

Section: Heterogeneity Of Logistics Requirementsmentioning

confidence: 99%

“…These methodologies are described shortly in the two following subsections. Piendl et al (2019Piendl et al ( , 2017 provide a more in-depth presentation.…”

Section: Modelling Frameworkmentioning

confidence: 99%

Section: Wider Implications Of the Transferability Analysismentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Building latent segments of goods to improve shipment size modelling: Confirmatory evidence from France

Piendl

Koning

Combes

et al. 2022

European Journal of Transport and Infrastructure Research

View full text Add to dashboard Cite

show abstract

“…Many modeling studies use an EOQ model as the theoretical basis for selecting the independent variables ( 3, 4, 9, 17–19 ), although the variables which are actually used in model estimation vary depending on data availability. Combes ( 2 ) assesses the validity of a conceptual EOQ model with theoretical coefficient values by comparing it with an empirical shipment size model estimated using domestic shipment records from the French ECHO database.…”

Section: Literature Reviewmentioning

confidence: 99%

Empirical Shipment Size Model for Urban Freight and its Implications

Sakai

Alho

Hyodo

et al. 2020

Transportation Research Record

View full text Add to dashboard Cite

Modeling shipment size for intra-city shipments is a subject that has not been sufficiently addressed in past research, despite its growing importance in disaggregate freight modeling. While past research on shipment size estimation mainly focuses on inter-city shipments, intra-city shipments differ from them in various aspects. In filling this research gap, this study estimates shipment size models using the records of intra-city shipments, identifying the effects of factors and heterogeneity on the shipment size selection mechanism. The estimated coefficients are also compared against their theoretical values derived from a conceptual economic order quantity model. The estimated empirical models highlight the characteristics of intra-city shipments and indicate the importance of both receiver function and commodity type, and also vehicle operation type, in capturing the nuances of the selection mechanism among intra-city shipments.

show abstract

Mode Choice Behaviour of Textile Shippers in India

Ansu

Anjaneyulu

2023

Lecture Notes in Civil Engineering

View full text Add to dashboard Cite

A machine learning approach for the operationalization of latent classes in a discrete shipment size choice model

Cited by 11 publications

References 12 publications

Building latent segments of goods to improve shipment size modelling: Confirmatory evidence from France

Building latent segments of goods to improve shipment size modelling: Confirmatory evidence from France

Empirical Shipment Size Model for Urban Freight and its Implications

Mode Choice Behaviour of Textile Shippers in India

Contact Info

Product

Resources

About