How it works
Software designed for the end-user
Senseye PdM is designed to be used on the shopfloor by the maintenance and operations teams who need to keep things running smoothly and ensure that unplanned downtime stays down.
The platform integrates seamlessly and provides maximum value by leveraging your existing investments. It focuses on automatically delivering advanced Predictive Maintenance insights in an easily understandable manner.
Condition monitoring data
Typically captured from sensors when machines are in a consistent operating state. Senseye PdM can also deal with operational parameters and more complex data sources including high-end condition monitoring hardware. If raw data requires transformation into usable condition monitoring data, this can be achieved using Senseye.
This provides Senseye with information on the operating context of an asset. Using operational data, Senseye can account for changes in machine behavior due to different workpieces, programs, or recipes.
Obtained from existing maintenance management systems or entered directly into the Senseye PdM app, this identifies maintenance activity or functional failures affecting a machine being monitored.
From basic MEMS vibration sensors to high-precision sophisticated Accelerometers, vibration monitoring gives a range of solutions for sensitivity and failure lead-time.
Changes in pressure and flow can indicate a number of different failure modes in process equipment. Something as simple as differential pressure across a pump can be used for PdM.
By taking data the control system is already capturing at a consistent state, modern drive units can capture torque readings that can indicate early signs of failure.
Information is collected directly from the PLC, requiring no additional hardware, or via unobtrusive retrofit sensors which enable legacy equipment monitoring.
Senseye PdM is developed for scale and is underpinned by a cloud-based platform capable of processing huge volumes of data – typically tens of thousands of related measures per hour. This enables you to reap the benefits of predictive maintenance at scale, applying it to every asset in every one of your facilities.
Unlike other solutions, Senseye PdM doesn’t require the development of custom models for each type of asset you operate. Custom models are typically restricted to critical assets due to the high cost of initial development. With Senseye PdM, models are constructed automatically with no user intervention – this means you can apply predictive maintenance to all of your assets, including lower criticality, covering the entire balance of plant.
Our data-driven approach means Senseye PdM can operate effectively with little or no context about the machines being monitored.
If available, however, contextual information such as production schedules, asset types, and asset criticality does help to enhance system output. Existing expert knowledge in the form of diagnostic rules and thresholds can also be utilized by Senseye PdM.
Match previous failure
Two main concepts underpin the analytics carried out by Senseye PdM:
Calculating the baseline fingerprint for each machine and detecting both isolated and ongoing deviations from this baseline. This process uses a fully automated, unsupervised approach.
Building a failure fingerprint for any historic functional failures and analyzing condition monitoring data to see if it matches a known fingerprint. This process is supervised, being triggered by the entry of maintenance data that identifies an actual functional failure.
A common challenge when you attempt to scale condition monitoring is the corresponding increase in user notifications. This challenge becomes even greater as you focus on applying predictive maintenance at scale across thousands of assets.
Senseye PdM uses a unique approach to direct your maintenance effort where it’s needed most. Central to this is the Attention Engine – a proprietary algorithm that estimates an Attention Index® for each of your assets, based on the recent maintenance data and patterns.
If the Attention Index® is sufficiently high, the Attention Engine creates a Case to direct the user’s attention to the asset in question.
The calculation of the Attention Index® factors in user feedback on previous patterns, asset criticality, and production schedule. The system learns and adapts to this user feedback and, as a result, generates more relevant notifications that are always prioritized consistently.
Putting the user experience first
Senseye PdM allows users to view and action cases, explore condition monitoring data for any connected asset, and report on maintenance activity.
Maintenance Engineers, Maintenance Managers, and Plant Managers benefit from the same clear, modern interface, usable on desktop and mobile devices.
A clear, prioritized list of Cases indicates assets that require attention. A simple workflow allows maintenance teams to acknowledge and track actions carried out in response to each case.
Engineers can explore condition monitoring data interactively on time-based plots, with access to the complete history of maintenance and previous cases.
IT and Operations
See at a glance the KPIs to track performance and calculate return on investment of Senseye PdM.
This includes the cumulative downtime avoided and current number of open/closed cases in the platform.
Plant & Maintenance Managers
Track and report on maintenance activity at any level, from a single asset to an entire production line or facility.
Explore maintenance effort and utilization to identify opportunities to improve efficiency.