How can legacy machinery deliver new business insights?
For many organizations, factories are pre-internet, 21st century assets. Machinery does its job day after day, often breaking down and needing attention. This can create an operational culture in which expectation and action are focused on that downtime, with maintenance teams repeatedly firefighting in between periods of normal service. The obvious alternative – investing in new equipment – is expensive, and installation would cause more even downtime.
Overall, manufacturing production can be efficient with old machines, but they do not offer holistic visibility across machinery performance and condition, which would help anticipate issues before they arise.
This conundrum was faced by a global consumer products company that had lots of manufacturing equipment running daily on legacy systems that were two decades old.
The equipment needed to do its job a little longer, but with higher expectations. These included making savings of $350 million over five years by analyzing operations and plant data to drive higher operational excellence, and to increase overall equipment effectiveness by 15%.
At the same time, the organization wanted to embrace Industry 4.0 and find a way to integrate siloed data across shopfloor systems into one using emerging technologies. Using artificial intelligence (AI), machine learning and data analytics would enable a remote, central view of performance that didn’t rely on only engineers inspecting equipment.
To that point, building predictive maintenance insights and proactive interventions was another high priority for the company. The cycle of machines running, unexpectedly failing and needing a maintenance crew to fix them led to a reactive, downtime-focused culture over the years.
Increased downtime meant decreased production. Something had to change.
Making machinery talk
To start the engagement, EY teams brought an internet of things (IoT) architect to dig deep into the client’s existing machines and systems, the problems they were facing and the best-suited solutions to fix them. The analysis identified two fundamental steps that were required to achieve the client’s objectives:
- Connect to local systems through a hardwired network.
- Deploy an on-premise edge solution to act as a middle layer between local systems and the cloud, which would effectively translate data into a common format that could be visualized in dashboards.
However, EY teams had to address two significant challenges. First, the machinery used several proprietary protocols, such as the OPC Unified Architecture (UA) protocol, and ways of communicating with machines that needed to be followed while using legacy software. This would require a bespoke IoT edge module in order to connect the machinery to the cloud. Second, the legacy systems were supplied by different vendors, which involved sending questions to vendors via the client and ultimately a complex analysis process.
To develop the solution effectively, EY teams involved the client shopfloor teams from the early stages. Maintenance crews wanted to gain machine-level insights by using different dashboards that would set out a procedure to follow and rectify faults quickly. EY teams designed these dashboards in consultation with the crews – the requirements came from them, in terms of widgets, key performance indicators (KPIs) and ease of adoption.
Using these insights, EY teams built a customized Smart Factory solution on the Microsoft Azure cloud platform. The complete shopfloor solution also used the EY organization’s IoT Sphere, which is a sensor-agnostic IoT cloud platform service. The solution helps accelerate technology to quickly generate asset monitoring and predictive analytics capabilities, along with an edge module to effectively translate the old-world machinery and legacy software data sources into cloud-compatible digital elements.
By helping to establish bidirectional, encrypted device communication and data ingestion services together, the solution would be able to compute and store data locally and display it with an intuitive dashboard. The dashboard would be pulled from an edge framework that would enable real-time data insights. Cybersecurity checkpoints were put in place too, so that data could be exchanged between machines and sites with a very low risk of breaches.
A factory that’s fit for the future
Working together with the EY teams, the client was able to reduce its unplanned downtime by 60%. The solution also drove a significant improvement in mean time between failures by an average of 126%, and overall equipment effectiveness increased by an average of 25% – well beyond the client’s stated operational objective ahead of the engagement.
The solution provides dashboards that the VP of Operations can use centrally to monitor performance, bottlenecks and production challenges. Now, IoT data helps predict maintenance.
Technology enablement unlocked by the Azure cloud solution also drove a change of culture. For example, capturing data from discrete sources on embedded sensors and controllers in the machinery, such as temperature, pressure and humidity, supports the solution to produce analytical models that provide recommendations for maintenance in scheduled downtime.
These insights have made work more straightforward for maintenance crews, and it has helped drive a change in organizational culture to be forward-looking, anticipatory and informed.
This engagement has helped EY teams refine the Azure cloud solution so that it can be offered to other clients too. There is tight integration across all frameworks that can be deployed on the edge and in the cloud, and additional modules for machine learning, digital twins and more. Each deployment will become more robust and offer more functionality. It will also lead to faster implementations, potentially including the client’s other sites.
This engagement shows that legacy operational technology can be made fit for the future without the vast expense of replacing it.
