Overcoming the three big challenges of IIOT data

blogOqtonNovember 6, 20185 minute read

Manufacturers we speak to say that they’re gathering industrial IoT (IIoT) data from their factories but aren’t necessarily getting the anticipated insights and benefits they expected from that data.

The challenges we see with factory implementation are that many existing IIoT solutions don’t monitor the necessary number of sensors per machine, don’t account for interactions between machines and fail to make full use of the data to report and produce insights that deliver on the promise of the technology.

Structuring systems and organizations to take full advantage of IIoT isn’t easy. The data lakes market is predicted to reach $9 billion by 2021. That will represent an immense amount of data stored and waiting for analysis. In a recent study, 74 percent of IT executives reported that the complexity of their data landscape limited their agility and 86 percent felt that they could be getting more from their data.

Further, most current technology applied to manufacturing data is reflective, leveraging data to understand what has already taken place in the factory environment. The industry is making advances in utilizing IIoT data for predictive maintenance of systems and equipment. We believe that the Holy Grail of IIoT will be the further actionable use of factory data in real time for planning, scheduling and programming of production.

But first, the industry will need to solve a few key challenges.

The Challenges of IIoT in Manufacturing

The falling cost of sensors has made IIoT data easier to gather from the factory environment, but manufacturers must overcome significant hurdles to ensure that they gather the right data and that they use it effectively to reduce costs and increase profitability.

1. Gaps in Data

Though manufacturing digitization is opening the door to increasing availability of data on the process, the number of sensors monitored per machine limits current IIoT solutions, leading to gaps in machine performance and status data that could otherwise be used to improve efficiency and throughput.

At a macro level, production technology remains highly heterogeneous. Not all workflow functions are inherently connected, generating gaps that can inhibit complete analysis and understanding. The very act of gathering data—particularly on work humans do—can bog down systems in programming and procedure. Closing these data gaps requires flexible solutions that enable agility and creative problem-solving.

2. Data Silos and Incompatible Formats

The fact that this data is coming from different machines and places in a variety of different formats is a related challenge. Manufacturers must also contend with data silos that contain the information they need to integrate and analyze. To understand the manufacturing environment, manufacturers need data from:

Machines
Quality inspection stations (human and robotic)
Inventory and warehouse functions (human and robotic)
Enterprise resource planning (ERP) and customer relationship management (CRM) system
The manufactured products themselves

The existence of these silos means that insights and potential improvements are also siloed. But production is a highly interdependent process; the success of each step depends on the outcomes of its predecessor and the requirements of its successor. To be truly effective, IIoT data needs to structure and rationalize this disparate data to provide insights on the interactions between machines, not just a machine alone.

Without a rational approach for gathering and structuring this data, it will stay in its silos, costing manufacturers money to maintain but greatly limiting its application and added value to the industry.

3. Insufficient Use of Data

The current state of the IIOT industry is that the technology to transmit the data has become achievable, but corresponding solutions to harness that data and make it useful are still nascent. Gartner estimates that one-third of IoT projects will be abandoned before deployment due to the lack of supporting data management and analytics capabilities.

A recent industry survey showed that while 77 percent of manufacturing organizations surveyed had adopting IIoT solutions to collect data, only 32 percent of them were automating even simple single-step actions such as support tickets from the data. This is far short of what the promise of the technology could deliver.

With high-speed and high-volume data aggregation and artificial intelligence (AI) to structure the data and make it useful and actionable in real time, the potential for IIoT-generated data is immense. Imagine the applications, particularly with advanced technologies such as additive manufacturing. Far more than simply alerting on status or predicting downtime, imagine pairing real-time data from the machine as it creates a build, with design and engineering data on what was planned. Now you have the capability to speed production time, reduce errors and greatly increase the reliability of additively produced production parts.

At Oqton, we see the greater potential of these technologies for the manufacturing industry and are focused on making that future a near-term reality.