Top Reasons your AVEVA/OSIsoft PI Implementation is Failing

The OSIsoft / AVEVE PI system is a powerful tool for industrial operations, providing valuable insights into data generated by industrial processes. However, despite its potential benefits, many organizations struggle to implement the system effectively. In this blog post, we will explore the top 11 reasons why OSIsoft / AVEVE PI system implementations may fail.

1. Incorrect System Differentiation: The PI system may have been added as an afterthought and lacks differentiated capabilities for trending, alerts and notifications, process visualization, kpi generation or analytics. Multi-site KPI generation may be impossible, making it difficult to compare sites effectively.

2. Data Coverage Issues: Under-subscribing to the number of tags / data streams the system has available can result in critical information being unavailable to users.

3. Bad Data Quality: IO timeouts, incorrect data compression strategies and other factors can result in bad data, leading to a lack of trust in the data.

4. Incorrect, Anemic Use of Analytics: PI-AF may not be leveraged properly to generate plant-level or inter-site KPI's, resulting in bare-bones information that leaves operators and engineers to their own devices.

5. Lack of Sponsorship: Operations leadership may not be bought-in to the use of the system, resulting in a lack of resources required to build out the system, including access to subject matter experts.

6. Lack of User-Level Training: Deploying the system without proper training for users can lead to confusion and lack of use.

7. Lack of Support Processes: Users may not know how to request support or report issues, and response times for OT/IT teams may lack Service Level Agreements (SLA).

8. Lack of Exception-Based Reporting / Alert Capabilities: PI notifications may not be used to proactively alert operations / engineering staff to problems, resulting in an ineffective use of domain expertise.

9. Lack of Thought Leadership: The organization may lack in-house expertise to prioritize and implement analytics capabilities on the PI system, resulting in a dated, anemic system that may not address the best use cases.

10. Lack of Cost Management: Over-subscribing to tags / data streams or client tools, or prematurely entering into an enterprise agreement, can result in unnecessary costs. In multi-site situations, opportunities to consolidate site historians into multi-site collectives & reduce tag /server costs may be missed.

11. Lack of Downstream Integrations into Enterprise Systems: Failing to leverage PI web services, RTQP or related technologies to automatically send correct, timely information into downstream systems like MES, maintenance systems, volumetric reporting tools, LIMS and related operations applications can lead to inefficient use of time and resources.

Rescuing a failed implementation of the OSIsoft / AVEVE PI system requires a comprehensive approach that addresses the root causes of the implementation failure. Here are some activities Lighthaus can help you undertake to rescue your system:

1. Conduct an Assessment: A comprehensive assessment of the implementation should be undertaken to identify the root causes of the failure. This assessment should look at the technical and non-technical factors that contributed to the failure.

2. Develop a Plan: Based on the assessment findings, a plan should be developed that outlines the corrective actions needed to address the identified issues. The plan should have clear timelines, roles and responsibilities, and measurable success metrics.

3. Address Technical Issues: Technical issues such as data quality, system differentiation, and analytics capabilities should be addressed. This can be done through implementing PI-AF and leveraging domain expertise to generate inter-site KPI's.

4. Address Non-Technical Issues: Non-technical issues such as lack of sponsorship, user-level training, and support processes should be addressed. This can be done through effective communication, training programs and service level agreements (SLAs) to ensure timely response to issues.

5. Develop a Governance Framework: A governance framework should be developed to ensure that the implementation is aligned with business objectives, and that roles and responsibilities for data management are clear.

6. Develop a Data Strategy: A data strategy should be developed that outlines how data will be collected, stored, and managed, and how it will be integrated with other enterprise systems.

7. Monitor and Measure Progress: The implementation should be closely monitored, and progress should be measured against success metrics. Any issues that arise should be promptly addressed.

By following these activities, organizations can rescue a failed OSIsoft / AVEVE PI system implementation and ensure that they are realizing the full potential of the system.