Visual Service Diagnostics and Service Provider Recommendation

Abstract - Home services management suffers from inefficiencies created by manual issue reporting, miscommunication, and delays in the allocation of service providers. Traditionally, the methods are reliant on users saying what the problems are, leading to wrong service assignments and increased time to resolve the problems. This paper presents an AI-supported Visual Service Diagnostics solution that entails automated issue discovery through Convolutional Neural Networks (CNN) and Machine Learning (ML)-based selection of service providers. Users upload images related to home service issues, which are analyzed to classify the problems into pre-defined categories such as plumbing, electrical problems, or appliance repairs. Then a ranking algorithm produces a recommendation of the most suitable service provider based on proximity to the problem, expertise, availability, and end-user ratings. The system runs on a seamless basis from Web and mobile client interfaces that allow the booking of services with real-time tracking and notifications. The solution maximizes efficiency through issue detection automation and service allocation optimization, hence minimizing user effort and maximizing customer satisfaction. Future extensions may include properties for voice diagnosis, AR support, and multilingual assistance for improved experience.