Talking about human-machine conversations in the industrial IoT

November 10, 2016

by



Tags: 

Share on FacebookShare on Google+Tweet about this on TwitterShare on LinkedIn

The “old way” humans interacted with machines in manufacturing largely depended on humans taking manual action in response to machine behavior. With the industrial internet of things (IIoT), a new way to interact with machines will increase automation, freeing the people in the conversation to focus on faster, smarter decision making made possible by machine intelligence.

blog-iiot-talking-about-human-machine-conversations_hero-800x247

Divvying up tasks between humans and machines is top of mind for industrial manufacturers. From the start of the 20th century, managing and operating complex industrial equipment has required “conversation” between humans and machines. Today, these conversations are increasingly complex. Industrial internet of things (IIoT) solutions that source data from industrial systems and analyze it to generate insights and influence operations are creating new opportunities to automate tasks that will increase the efficiency of human workers. But for industrial manufacturers to take full advantage of the IIoT, they need to understand how the conversation between humans and machines—including the language itself—is evolving with the advent of the IIoT (see graphic below).


The future of human-machine dialogue

iiot-future-human-machine-dialogue

IIoT solutions will expand the conversations in the industrial ecosystem from the two-way conversation that has evolved over the years to three-way conversations in the future, creating new opportunities to improve.

The evolution of human conversation with industrial equipment

In the early days of managing manufacturing equipment, the conversation consisted of signals and senses. The sight, smell, sounds, and touch of the equipment relayed important information that experts used to judge if things were fine or if some maintenance or operational change was needed. Steam coming out of the cooling system, for instance, might have been an indication that something was too hot.

In the next generation of equipment, sensors built into the machines could measure temperature, pressure, current, and so on, and display it on dials. The conversation extended to the internal state of the equipment, such as oil pressure, engine temperature, and speedometers in automobiles. Any intervention or response still required human action.

In the subsequent generation of equipment, the addition of feedback control systems enabled a certain level of automation. For instance, if a piece of equipment was overheating, automation feedback could trigger cold water to be added to the cooling system in order to reduce the temperature. But the number of sensors and the ability to communicate, share, or combine the information surfaced by the sensors was limited.

Today with IIoT systems, sensors are proliferating both inside the equipment and in the environment in which it operates. Data coming from these sensors is much more detailed—and it’s generated continuously. With the ubiquity of the internet, this data can be transported, shared, combined, and analyzed—opening doors to new conversations and ways to make systems more intelligent. Previously, 100 percent of the conversations took place between humans and industrial equipment. The future will look a lot different—more like the graphic below, which shows the conversation between the different participants.


Deriving value from the IIoT conversation

iiot_blog_human-machine-conversations2

Shifting interactions in the industrial ecosystem suggest where new value is being created with the adoption of the IIoT. The size of the areas in this illustration roughly suggests the size of each opportunity.

Which IIoT conversations will create the most value?

Here are the possible conversation changes—and their potential benefits to companies:

  • Area A represents the human-to-machine conversation that is still a largely manual operation. For instance, someone sees that the pressure is too high on a gauge and turns a knob to reduce it. The frequency of interactions and the value have been here for a long time.
  • Area B represents the human-machine conversation. This area will be smaller than A, because IIoT solutions relieve humans of a number of manual supervision, monitoring, and inspection tasks. The volume of interactions and the value in this area is likely to decrease.
  • Area C represents machine-to-IIoT conversation. The bulk of this conversation consists of data flowing to the IIoT system. The data will be analyzed—by using machine learning and artificial intelligence techniques—to identify noteworthy correlations and patterns. This conversation includes opportunities for automation where the IIoT solution can provide instructions to machines without human intervention—slowing down an engine if it is overheating, for instance, or other actions to keep systems performing optimally. Interactions in this area will be more frequent and will create new value.
  • Area D represents IIoT-to-human conversation. IIoT solutions will filter out the noise and present only pertinent signals in the form of patterns, alerts, notifications, and correlations to humans, who will provide validation, feedback, decision, or action. This is another large area of new value creation.
  • Area E represents conversations among humans, machines, and IIoT solutions. These are manual insights that can be validated by IIoT analysis, increasing human confidence in the insights and perhaps changing the way operators learn about the behavior of the equipment.

The type and frequency of these IIoT-driven conversations are good indicators of where the IIoT will create new value. The IIoT will expand humans’ visibility into the behavior and effectiveness of the industrial machines and equipment. Human involvement is not likely to be eliminated completely, but the nature of information that people handle and the tasks they must perform will become a combination of manual interaction, early notification of important events (failures, potential disruptions, inefficiencies, and so on), ideas for process improvements, and decision support.

Are you seeing these conversation changes with the adoption of the IIoT? How are you managing the transition?

Share on FacebookShare on Google+Tweet about this on TwitterShare on LinkedIn

Industries

Contacts

Chris Curran

Principal and Chief Technologist, PwC US Tel: +1 (214) 754 5055 Email

Vicki Huff Eckert

Global New Business & Innovation Leader Tel: +1 (650) 387 4956 Email

Mark McCaffery

US Technology, Media and Telecommunications (TMT) Leader Tel: +1 (408) 817 4199 Email