Even if the Plant is well represented in plant layout, and logistic warehouse is tracked on daily bases, during a normal shift of drivelines, the production lines are suffering unplanned and planned stoppages that are affecting their availability; such stoppages are typically requiring to replace worn components (e.g. tooling), to maintain manufacturing conditions, such as cleaning the machine of the chips produced during the process, or are caused by components malfunction (e.g. broken tools, inductive switches, electro valve). Another important issue is the performance loss and the final quality of the product, which is caused by faulty components (not perfectly in line with 3d-2d models), poor lubrication, bad machine setup, etc. and results in an increased machine cycle time, affecting the entire line as lines are configured as mass production (one operation feeds the next operation).

With the help of the XMANAI AI platform, CNHi will ingest, manage and analyse the real-time and batch data acquired by CNHi systems (manufacturing, production, logistic, trace data), and the scheduled data contained in the maintenance and Tooling systems, in order to build novel AI models that contribute to the provision of recommendations to optimize the line throughout of the current and successive shifts. The XMANAI demonstrator will take actual data from the current systems that brings information about these different parts/and how they have been assembled along the line. In such a way, the demonstrator will test how to create a more organized data management, sharing, and generate knowledge graphs with clear relationships between data and defined actions. The hybrid and graph AI models to be explored, trained and evaluated in XMANAI will allow for simulating different scenarios related to optimizations and predictions for: (a) the production batches in terms of size, mix and schedule basing on 3d-2d models, (b) the tool change strategy based on line efficiency and tool life, and (c) maintenance tasks based on line efficiency, machine/people/tools availability and considering the last observed machine behaviour (failure mode elimination or cycle time reduction). The demonstrator will take advantage of the use of virtual reality tools to represent new manufacturing scenarios that will help operators by visualizing and explaining the ΑΙ outcomes in natural language, promoting understanding and trust from users, as well as inspection and traceability of actions undertaken by the AI systems. The actionable results of the AI algorithms will be leveraged in the CNHi-specific applications that will be developed in XMANAI in order to communicate to operators intelligent automatic alerts and recommendations. In this way, it will be possible to identify at any given moment, the reason/s which are jeopardizing production and take appropriate, informed action using system level digital twin which uses AI. By testing nominal and abnormal scenarios (failures, disruptions in the production, etc.), the demonstrator will allow to explore how to facilitate a correct and timely response by the operators without overloading them, thus mitigating the risk of error and increasing trust and acceptance.