Share

IIR document

Ageing of refrigerated transport vehicles: development of a numerical predictive model.

Number: 294815

Author(s) : CAPO C., LE GUILLY M., PAPE L., SCUTURICI M., PETIT J. M., REVELLIN R., BONJOUR J., CAVALIER G.

Summary

Ageing is a common problem for refrigerated vehicles. It involves both physio-chemical and mechanical factors. This phenomenon is quantified by the comparison of the 𝐾 value of the in-service vehicles at twelve years of use with the 𝐾 value of their prototypes. The Datafrig® database, managed by Cemafroid, records more than 350 000 vehicles having almost one hundred features, including 𝐾 coefficient values. This amount of data opens up the opportunity to predict the ageing by building a numerical model, using supervised machine learning techniques. This paper presents the development of a model using a random forest algorithm. Results based on a first dataset, composed of 1158 vehicles extracted from the Datafrig® database, agree with the knowledge of the experts from field experience as well as with previous studies based on statistical analysis. After a careful feature selection, the mean percentage error reached by the model for this first dataset is estimated at 6%.

Available documents

Paper

Pages: 8

Available

  • Public price

    20 €

  • Member price*

    Free

* Best rate depending on membership category (see the detailed benefits of individual and corporate memberships).

PowerPoint version

Pages: 16

Available

  • Public price

    5 €

  • Member price*

    Free

* Best rate depending on membership category (see the detailed benefits of individual and corporate memberships).

Details

  • Original title: Ageing of refrigerated transport vehicles: development of a numerical predictive model.
  • Record ID : 30027606
  • Languages: English
  • Subject: Technology
  • Source: 6th IIR International Conference on Sustainability and the Cold Chain. Proceedings: Nantes, France, August 26-28 2020
  • Publication date: 2020/08/26
  • DOI: http://dx.doi.org/10.18462/iir.iccc.2020.294815
  • Document available for consultation in the library of the IIR headquarters only.

Links


See other articles from the proceedings (84)
See the conference proceedings

Indexing