Manufacturers face tough decisions that impact their ability to grow their business and increase profits because of unexpected disruptions in production due to equipment failures. A critical requirement is finding an optimal strategy to keep their equipment in peak condition for maximum asset availability while minimizing costs. Prioritizing when and what to maintain is a complicated problem that is often not adequately addressed because of a lack of information from both business and execution systems.
In manufacturing, maintaining a machine too often can be as bad as maintaining it too little. The previous statement may seem unintuitive, but preventive maintenance based on a schedule can be costly because one is changing parts and tooling that still have a useful life. It also ties up working capital for planned-downtime that may not be necessary.
When one does not perform adequate service, the parts will wear and failure will be unpredictable and can be catastrophic. The situation compounds if there is no way to know precisely how much time and under what conditions the equipment was used.
For a maintenance engineer, it is not apparent where to best apply one’s energy to maximize production and profit. The engineer must know the state of every piece of equipment, the job schedules, and the contractual deadlines to make a holistic assessment of the current tickets. The information is sometimes available, such as the current job schedule, but it is often not available in a form that is readily consumable by maintenance. Add to this the need to perform scheduled maintenance and the logistics often become untenable.
It is for this reason that many maintenance engineers can never get ahead and are always addressing crises instead of being able to schedule their time effectively.
VIMANA addresses these problems by capturing the precise history of a piece of equipment and integrates the ecosystem of business and execution systems to find an optimal strategy. The usage and history of the machine are achieved by collecting data to understand the processes and the operational conditions under which it was used.
The history is analyzed to find patterns that are subsequently applied to the real-time data streaming from the control systems. The patterns provide insight into how the machine is performing and when preventative maintenance should be performed based on the usage.
Integration of the machine tool and real-time execution data with the business systems allows the maintenance engineer to understand where critical jobs are running and the business impact of an unexpected disruption. The integration also allows for business process automation by enabling the equipment to automatically enter tickets when it detects anomalous behavior and associates it with the jobs, contracts, and the historical record.
As the history and ecosystem integrations progress, the analytics become more prescriptive, allowing the business to predict the impact of a job on a machine and the current capabilities to perform a given set of operations. Prescriptive analytics are the ultimate goal since when one can accurately predict the impact and abilities, unexpected disruptions become more infrequent and ticket prioritization is automated.
