Preventive maintenance for heterogeneous industrial vehicles with incomplete usage data

Abstract Large fleets of industrial and construction vehicles require periodic maintenance activities. Scheduling these operations is potentially challenging because the optimal timeline depends on the vehicle characteristics and usage. This paper studies a real industrial case study, where a company providing telematics services supports fleet managers in scheduling maintenance operations of about 2000 construction vehicles of various types. The heterogeneity of the fleet and the availability of historical data fosters the use of data-driven solutions based on machine learning techniques. The paper addresses the learning of per-vehicle predictors aimed at forecasting the next-day utilisation level and the remaining time until the next maintenance. We explore the performance of both linear and non-liner models, showing that machine learning models are able to capture the underlying trends describing non-stationary vehicle usage patterns. We also explicitly consider the lack of data for vehicles that have been recently added to the fleet. Results show that the availability of even a limited portion of past utilisation levels enables the identification of vehicles with similar usage trends and the opportunistic reuse of their historical data.

[1]  Octavian Morariu,et al.  Machine learning for predictive scheduling and resource allocation in large scale manufacturing systems , 2020, Comput. Ind..

[2]  Mohammed J. Zaki,et al.  Data Mining and Machine Learning: Fundamental Concepts and Algorithms , 2020 .

[3]  Paolo Santi,et al.  Characterizing the "Driver DNA" Through CAN Bus Data Analysis , 2017, CarSys@MobiCom.

[4]  Henrik Almer Machine learning and statistical analysis in fuel consumption prediction for heavy vehicles , 2015 .

[5]  Mehdi Rashidnejad,et al.  A bi-objective model of preventive maintenance planning in distributed systems considering vehicle routing problem , 2018, Comput. Ind. Eng..

[6]  Elke A. Rundensteiner,et al.  Generalized Dynamic Time Warping: Unleashing the Warping Power Hidden in Point-Wise Distances , 2018, 2018 IEEE 34th International Conference on Data Engineering (ICDE).

[7]  Zahid Halim,et al.  Profiling drivers based on driver dependent vehicle driving features , 2015, Applied Intelligence.

[8]  Jing Zhang,et al.  CAD: Command-Level Anomaly Detection for Vehicle-Road Collaborative Charging Network , 2019, IEEE Access.

[9]  Eamonn J. Keogh,et al.  Exact Discovery of Time Series Motifs , 2009, SDM.

[10]  Hongchao Wang,et al.  Data-Driven Methods for Predictive Maintenance of Industrial Equipment: A Survey , 2019, IEEE Systems Journal.

[11]  Karl Henrik Johansson,et al.  Vehicle Applications of Controller Area Network , 2005, Handbook of Networked and Embedded Control Systems.

[12]  Nigel N. Clark,et al.  Heavy Duty Truck Fuel Consumption Prediction Based on Driving Cycle Properties , 2012 .

[13]  Daniela M. Witten,et al.  An Introduction to Statistical Learning: with Applications in R , 2013 .

[14]  Erik Hellström,et al.  Look-ahead Control for Heavy Trucks to minimize Trip Time and Fuel Consumption , 2007 .

[15]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[16]  Teodor Gabriel Crainic,et al.  Fleet management and logistics , 1998 .

[17]  Luca Cagliero,et al.  Heterogeneous Industrial Vehicle Usage Predictions: A Real Case , 2019, EDBT/ICDT Workshops.

[18]  Álvaro Herrero,et al.  Studying Road Transportation Demand in the Spanish Industrial Sector Through k-Means Clustering , 2018, SOCO-CISIS-ICEUTE.

[19]  Luís C. Neves,et al.  Application of machine learning for fuel consumption modelling of trucks , 2017, 2017 IEEE International Conference on Big Data (Big Data).

[20]  Simaan M. AbouRizk,et al.  Data-driven simulation-based model for planning roadway operation and maintenance projects , 2017, 2017 Winter Simulation Conference (WSC).

[21]  Mehmet Sevkli,et al.  Fuel Consumption Models Applied to Automobiles Using Real-time Data: A Comparison of Statistical Models , 2016, ANT/SEIT.

[22]  Takayuki Morikawa,et al.  Exploring Trip Fuel Consumption by Machine Learning from GPS and CAN Bus Data , 2015 .

[23]  Wenshuo Wang,et al.  A review of vehicle fuel consumption models to evaluate eco-driving and eco-routing , 2016 .

[24]  Luca Cagliero,et al.  Machine Learning Supported Next-Maintenance Prediction for Industrial Vehicles , 2020, EDBT/ICDT Workshops.

[25]  Dimitrios Gunopulos,et al.  Mining Time Series Data , 2005, Data Mining and Knowledge Discovery Handbook.

[26]  Sandareka Wickramanayake,et al.  Fuel consumption prediction of fleet vehicles using Machine Learning: A comparative study , 2016, 2016 Moratuwa Engineering Research Conference (MERCon).

[27]  Jorge Luis Victória Barbosa,et al.  Machine learning and reasoning for predictive maintenance in Industry 4.0: Current status and challenges , 2020, Comput. Ind..

[28]  Christophe Bérenguer,et al.  Joint dynamic scheduling of missions and maintenance for a commercial heavy vehicle: value of on-line information , 2018 .

[29]  T. Pereira,et al.  Industry 4.0 implications in logistics: an overview , 2017 .

[30]  Thuy T T Nguyen,et al.  Fuel consumption estimation for kerbside municipal solid waste (MSW) collection activities , 2010, Waste management & research : the journal of the International Solid Wastes and Public Cleansing Association, ISWA.

[31]  Eamonn J. Keogh,et al.  Matrix Profile X: VALMOD - Scalable Discovery of Variable-Length Motifs in Data Series , 2018, SIGMOD Conference.