Artificial Intelligence & Machine Learning
Predictive ML, the SaMi AI assistant and IoT data – working on your CMMS and production data, in the cloud or on-premise.
SimplyMobile combines classic machine learning (prediction) with generative AI (large language models) and IoT device data into one coherent AI layer over your CMMS and production data. You can deploy all of it in the Azure cloud or on local, on-premise servers, keeping full control of your data.
SaMi – the AI assistant that chats with your data
SaMi is a built-in conversational assistant powered by Azure OpenAI, with the option to connect other models as well. You ask questions about your data in natural language – work orders, assets, inventory, readings – and get answers, summaries and insights, without building reports.
- Natural-language questions and answers about your data
- Powered by Azure OpenAI, with the option to connect other models (model-agnostic)
- Works across your CMMS and production data
- Your data stays in your environment – cloud or on-premise
Talking to your data through MCP
SaMi uses an MCP (Model Context Protocol) server built into the application, which securely exposes your operational data to language models. This lets the AI reason over real CMMS and production data – opening up advanced production and maintenance scenarios driven through conversation.
Predictive maintenance – Simply ML
The Simply ML module backs decision-making and diagnostic processes with advanced computational tools. Introducing prediction helps optimize equipment usage and prevent some failures before they stop production.
Key components of the Simply ML module
- ML Configuration Module
- Data Aggregation Module
- Prediction Module
- Reporting and Alerts Module
Simply ML implementation process
- Identify the equipment subject to prediction
- Install new sensors or integrate with existing ones
- Identify event types and map them to historical data
- Prepare data and train predictive models
- Deploy, monitor, and optimize predictive models
Meter readings and IoT data
A dedicated platform for meter readings (IoT) feeds the system with device data in near real time. This is the foundation for prediction, condition-based maintenance and reliable metrics.
Training and learning your own models
Beyond ready-made predictive models, the platform lets you train and continuously improve models on your own historical data – tailored to the specifics of your plant.
Predictions that support everyday decisions
The AI can be trained on your historical data to suggest and forecast values, speeding up routine decisions. Examples:
- Suggest who should approve a given document (approval routing)
- Propose the accounting posting (decree) to generate for a document
- Estimate optimal inventory levels (min/max, reorder point)
- Forecast equipment failures before they happen
Deployment: cloud or on-premise
We deploy the entire AI layer – Simply ML, the SaMi assistant and the MCP server – in the Azure cloud or on the client's local servers. This matters wherever data must stay within the plant's own infrastructure.