Event Overview: At this event we were joined by 100+ participants from 40+ operators to discuss the issue of sensor data quality, both the impacts of poor sensor data quality as well as how Aperio helps to solve the issue. In addition to the presentation from Aperio, we were lucky enough to be joined by Jane Arnold, VP of Global SPS and Industry 3.0/4.0 with Stanley Black & Decker, who is an expert in the space of digital transformation and has worked with Aperio in the past.

Over the past 5-10 years the industry has pushed to increase the amount of data collected in the field. This has led to material improvements in operational efficiency, real-time visibility into the status of operations and the rapid development of machine learning/optimization software solutions. While the massive increase is sensors and other connected devices in the field has been a net-positive, the old adage of “garbage in, garbage out” still applies.

When we discuss solutions within the digital space, we group vendors into one of four buckets – acquisition, data management, analytics and automation – which broadly follow the life of a byte of data. The first step is to acquire the data, then it has to be managed, once it is managed then we can start developing analytics and automation solutions on top of it. It is critical to bear-in-mind that these steps are cumulative, meaning you can’t jump to automation without either collecting, managing, or analyzing the data available.

Where solutions like Aperio come into play and can add substantial value is in this data acquisition and management portion of the life cycle. While this is a critical step in the process, it is often very a manual procedure where many of our engineers and data teams are spending the bulk of their time manually adjusting and cleansing the data. Which means that they can’t spend their time and energy in potentially higher value efforts around analytics and automation initiatives. In fact, 35% of respondents to the question of how they are currently addressing sensor data quality, only 10% have a dedicated solution while 35% rely on manual intervention.

For further details on the value of and challenges associated with ensuring high quality sensor readings, please take a minute to watch this presentation from Jane Arnold of Stanley Black & Decker.

The bulk of the Q&A during the event was centered around three key themes:

• Q: How is the Aperio solution connect to existing architecture/workflows?

• A: The software can be connected via either API or deployed in real-time and is able to ingest and cleanse the data in near real-time which can then be written back to the historian.

• Q: How does the system account for sensor drift or potentially erroneous readings from the beginning (i.e. ensure that you aren’t using a faulty baseline)?

• A: In addition to reading the baseline, Aperio has integrated a signal processing solution which can help to identify poor readings and detect any drift over time which would potentially be missed by solutions only looking for outlier events.

• Q: How can this data quality score be integrated into a maintenance or workflow solution that automatically calls out repairs in the field?

• A: This is a trickier solution as it requires the integration of ERP solutions, dispatching, etc while also addressing the change management impact of a new workflow for the organization. In addition to minimizing the aforementioned manual time and energy spent on data cleansing, most of the attendees (and confirmed by Jane Arnold) the value of Aperio and similar solutions lies in improved analytics solutions (41% of respondents) and in limiting the number of false positives which lead to an inordinate amount of downtime associated with alarming. The other area where value is created, yet is less visible, is that by minimizing the number of false alarms, operators are able to decrease alarm fatigue which can lead to potential HSE events.