J. Valero et al., "A Statistical Size Recommender for Safety Footwear Based on 3D Foot Data", Proc. of 3DBODY.TECH 2023 - 14th Int. Conf. and Exh. on 3D Body Scanning and Processing Technologies, Lugano, Switzerland, 17-18 Oct. 2023, #40, https://doi.org/10.15221/23.40.

Title:

A Statistical Size Recommender for Safety Footwear Based on 3D Foot Data

Authors:

Jorge VALERO 1, Alfredo BALLESTER 1, Beatriz NACHER 1, Clara SOLVES 1, Cataldo DE LUCA 2, Marta PESCE 2, Sandra ALEMANY 1

1 Instituto de Biomecanica de Valencia IBV, Valencia, Spain;
2 Base Protection S.r.l., Barletta, Italy

Abstract:

Shoe size recommendation remains a significant challenge for the footwear industry. Getting a shoe that does not fit leads to customer dissatisfaction and high return rates. In the case of safety footwear, this challenge is even greater because wearing the wrong size can compromise the safety of workers. It is thus crucial to develop a technology that allows industry to efficiently provide their employees with the right shoe size advice, in a fast, simple, and effective manner. This paper describes the results of the joint cooperation between IBV and Base Protection S.r.l. to develop and deliver such a system.
The proposed technology uses low-cost 3D scanning technology (Domescan/IBV and 3Davatar feet) to accurately capture 20 foot features and a shoes size recommender based on Multinomial Logistic Regression (MLR). The system was trained for 14 shoe lasts and used data from fitting tests of 60 subjects from both genders. It was validated with fitting tests of 25 subjects achieving an 60-80% success rate in recommendations, depending on the shoe model. The results also showed that personal fit preference plays a crucial role in size selection, hindering greater accuracy. In this regard, one of the main advantages of MLR is its informative output, i.e. a map of fitting probabilities for each size, which offers multiple options for the development of the user interface layer and may enable that the final consumer to make an informed decision based on it. The system also included an insole recommender (low, mid and high arch) that uses a classic two-dimensional recommendation grid based on foot arch indexes based on three foot features. These technologies were embodied into a physical booth for brick-and-mortar stores and into an app that directs the consumer to the nearest point of sale. This system represents a significant advance in the footwear industry and offers a streamlined solution for brands and retailers. Overall, this work demonstrates the effectiveness of utilizing MLR in a footwear recommender system, and highlights its potential for footwear brands and retailers to reduce returns and increase sales, for consumers to get a better comfort and safety at work, and for industries needing for safety shoes to reduce the burden of managing the footwear orders to its employees.

Keywords:

safety footwear, shoe fitting, size recommendation, 3D foot, logistic regression, anthropometry

Details/PDF/VIDEO:

Full paper: PDF
Presentation: VIDEO
Proceedings: 3DBODY.TECH 2023, 17-18 Oct. 2023, Lugano, Switzerland
Paper id#: 40
DOI: 10.15221/23.40

Copyright notice

© Hometrica Consulting - Dr. Nicola D'Apuzzo, Switzerland, hometrica.ch.
Reproduction of the proceedings or any parts thereof (excluding short quotations for the use in the preparation of reviews and technical and scientific papers) may be made only after obtaining the specific approval of the publisher. The papers appearing in the proceedings reflect the author's opinions. Their inclusion in these publications does not necessary constitute endorsement by the editor or by the publisher. Authors retain all rights to individual papers.

Note: click the + on the top left of the page to open/close the menu.