Context-Based Maintenance and Repair Shop Suggestion for a Moving Vehicle

Maintenance of moving vehicles is quite challenging because they may disrupt the normal flow of transportation due to unexpected breakdowns, slowdowns and stoppages. In order to avoid stoppages and to minimize the downtime, maintenance and condition monitoring systems must be optimized. On one hand the condition monitoring on board should provide automatic failure detection, identification and localization together with a prognostic of the future failures. On the other hand maintenance logistics and product supportability must be also optimized since the onboard system should provide a suggestion of a repair shop that depends on location, cost and availability of spare parts, technicians’ skills and queuing time for repairs. However the vehicles are independent assets interacting among them within the traffic system and also interacting with the infrastructure (roads, rails etc.) seriously affected by weather, maintenance of infra, regulations etc. Therefore the proposed solution is to equip the vehicles with a context-aware system that monitors the condition and maintenance schedules of parts and alarm the driver of the parts that are in near to repair cycle. This system will perform risk analysis and will communicate with the cloud propose a decision of selection of repair shop on the location and path of vehicle depending on weather, road and traffic, cost and availability of spare parts at respective repair shops based on risk assessment and prediction. The information contained in the cloud will also communicate the workshop that will book time slot and block the necessary spare parts for the coming vehicle minimizing waiting time. This mechanism will help in reducing unexpected stoppages, vehicle degradation and efficient spare parts management combining in a successful way the workload of the workshops from both natural sources, the time based inspections and repairs together with the reactive maintenance coming from unexpected breakdown.