{"slug":"totalenergies-ai-agent-marge-improves-fault-cause-identification-25-45-in-industrial-maintenance","url":"https://findausecase.com/use-cases/totalenergies-ai-agent-marge-improves-fault-cause-identification-25-45-in-industrial-maintenance","title":"TotalEnergies' AI agent MARGE improves fault-cause identification 25-45% in industrial maintenance","description":"TotalEnergies, a global energy producer operating in nearly 120 countries, rolled out 30,000 Microsoft 365 Copilot licenses to employees and built autonomous AI agents on Microsoft Copilot Studio. MARGE (Machine Assisted Reliability and Graving Engine) automatically analyzes and classifies industrial maintenance intervention reports that were previously handled manually, improving the identification of fault causes by 25% to 45%. A second agent, BuyerCompanion, built with partner Witivio, streamlines procurement for transactions up to €50,000 by drafting specifications, identifying suppliers and relevant framework agreements, and generating personalized recommendations, with an estimated 10% savings on certain purchases; it is already operational in France and could ultimately support 15,000 staff worldwide.","company":"TotalEnergies","industry":"Energy","country":"France","aiCapabilities":["Agentic AI","Document Intelligence","Generative AI"],"technology":["Microsoft Copilot Studio","Microsoft 365 Copilot"],"deployment":"Unknown","problemStatement":"TotalEnergies' industrial maintenance intervention reports were previously analyzed and classified manually, and its procurement processes for smaller transactions required significant manual work drafting specifications and identifying suppliers.","solutionApproach":"TotalEnergies built MARGE (Machine Assisted Reliability and Graving Engine), an agent based on Microsoft Copilot Studio deployed with support from Microsoft Industry Solutions teams, to automatically analyze and classify intervention reports, and BuyerCompanion, built with partner Witivio using Microsoft Copilot Studio, to streamline procurement processes for transactions up to €50,000.","businessValue":"MARGE improved the identification of fault causes by 25% to 45% compared to the previous manual process, while BuyerCompanion is estimated to save 10% on certain purchases and, already operational in France, could ultimately support 15,000 staff worldwide.","evidence":{"band":"high"},"sourceUrl":"https://www.microsoft.com/en/customers/story/25505-totalenergies-agents","dates":{"publishedAt":"2026-09-15T09:05:33.056Z","publishedAtSource":"pipeline","updatedAt":"2026-09-15T09:05:33.056Z"},"license":"Open for reading and citing with a link to https://findausecase.com/use-cases/totalenergies-ai-agent-marge-improves-fault-cause-identification-25-45-in-industrial-maintenance. Bulk republication requires permission."}