Digital Twin For Asset Management And Predictive Maintenance

Digital Twin for asset management and predictive maintenance through digitization

A global Cement Producer started a digital transformation journey to increase revenue. This venture is set to reduce production downtime, improve efficiency by more than 10%, and reduce CO2 emissions. The key to this transformation is the creation of a Digital Twin using data provided by the company’s many assets.

Using Wizata’s machine learning models, the project returned +€2 million in savings over the first two years. It was also delivered on time and on budget. Wizata also trained the company's local resources to maintain, manage and evovle the system.

Priority in the upstream cement industry is to shift towards environmentally friendly production. Plants face pressure to reduce downtime, maximize margins and improve health and minimize environmental impact.

Solution to these challenges had the potential to improve two key business areas in phase l of the transformation:

Real-Time Anomaly Applications
Data was present in the historian database but lacked contextualization and flexibility. The treatment also disallowed leveraging of data. Simple monitoring and anomaly detection models need to be applied to real-time telemetry data, but the existing infrastructure does not allow the execution of new real-time calculations.
Predictive and prescriptive maintenance on key assets like Motors, Gearboxes, Pumps, and ID Fans to reduce and predict time to failure. The empirical programmed maintenance shifted to Al-based prescriptive maintenance.

Wizata and the local experts worked closely with business leaders to ensure that the project was aimed at producing the biggest business impacts and generating measurable ROI.

The following activities were performed:

The Wizata platform, hosted within customers’ Azure tenant
Ingest metadata from OSlsoft Pl, MES, Historians, and engineering data sheets
Design and build a process Centric Digital Twin based upon the target equipment
Transform and contextualize process and maintenance data
Use the embedded ML models of the platform and produce real-time recommendations
Build customized dashboards for relevant stakeholders

RESULT

Wizata is demonstrating the power of asset data modeling and replication at scale. With the implementation, the company is experiencing a 3% reduction in energy consumption. CO2 emissions have also been reduced. Unexpected stoppages have decreased by 25%.

The Wizata Anomaly Detection system is currently running 16 models onto 35 assets included in the first phase. The scalability of the whole solution is facilitated by digitalizing each production process through Digital Twin and applying replication templates to adapt, deploy and maintain intelligence algorithms within the group.

The contextualized data within Wizata now allows for quick responses to incidents in real-time to prevent unplanned downtime and improve reliability.

USE CASE 02
CEMENT INDUSTRY