Predictive scrap solution is part of the add.AI solution portfolio which leverage power of Artificial intelligence to improve operation and business.

Key Benefits of the solutions:

Predict scrap rate and understand key factors for scrap generation.
Improve quality and lower rework and defected products.
Environmental benefits: fewer defects, more minor (mis)usage of resources (workforce, energy, raw material). It is a good starting point for transformation to become a green Manufacturer.
Increase productivity and profitability.
Alerting and notifications: make corrective actions and prevent scrap.
Analyze and empower your employees to take smart decisions.

How it works:
Predictive Scrap solution is built with the Machine Learning technology to learn from historical data and real time data to analyze failure patterns. Since conservative procedures result in resource wastage, Predictive Maintenance using Machine Learning looks for optimum resource utilization and predicting failure before they happen, on the machine level (predictive maintenance) and on the product level (predictive scrap).

About solution:
Solution will be cloud-native and includes following Azure services:

Azure IoT Hub: To provide a cloud-hosted solution back end to connect virtually to any device. Service is needed for capturing real time data. It offers possibility of registration machines and/or sensors in production in cloud as individual devices (eg. A machine or several machines are registered as a Device in Azure IoT Hub serves as client for data transmissions generated on the machines).
Azure Stream Analytics: To provide real time parsing of stream data (transformation, aggregation of data).
Azure Data Lake for long term storage of data (cold storage of data).
Azure Synapse for dimensional models’ creation and reporting for non-real time data which can enriched with other data from production. Presentation layer in Power BI.
Power BI real time analytics and alerting (presentation layer and analytics).
Azure Databricks: For predictive models learning and near real-time forecasting. Proposed solution anticipates forecasting in different time intervals (real-time and/or 5 minutes and/or 30 minutes and/or 1 hour).
Supporting services (Azure AD, Azure Key Vault, Azure Logic Apps, Azure Functions, Azure DevOps, Azure Monitor) are used to ensure security, orchestration, and monitoring of Azure services utilization.

Timeline- Agenda:

1st - 3rd week: Envisioning Workshops to identify data sources required for AI model. GAP-FIT Assessment
4th - 8th week: Data integration and deployment of AI model (CRISP methodology)
9th - 10th week: Stabilization & final revision with the customer, model & results presentation

Payment model: SaaS

Monthly Fee = FIX (predictive scrap solution fee * number of machines) + FLEX (Azure Services).
Price and Scope will vary depending on the requirements, size, and complexity of your environment.

Learn more about us and our solutions for Manufacturing companies and our add.AI product portfolio.

Azure Integration Workshop - 2 Stunden

DATA Passion GmbH

Details

Links