The very concept of automation leads to a moral and ethical dilemma regarding the workforce. People need to work to support their families, but there are tasks that could be done better, faster, and cheaper if automated. This leads to a tension between upper management and employees where one group is actively interested in automation and the other group is actively threatened by it. This two-sided debate is the one that gets the most airtime, but what doesn’t get covered is the win-win resourcing conversation where employees are freed up from doing the menial tasks that sensors and data loggers are best at, allowing the employees to do higher-value-adding tasks that require interaction and thought. By being freed up to focus on these types of tasks employees are able to hone their personal craft, creating a higher level of intrinsic job satisfaction.
Using a (highly) abridged version of the Job Characteristic Theory framework (Hackman 1975) we can see that through applying automation via the Internet of Things (IoT) there will be an actual increase employee satisfaction.
| Core Job Characteristic |
Description of Characteristic |
IoT Solution |
Employee Impact |
| Skill Variety |
More diverse skills used and fewer rote tasks performed lead to higher satisfaction |
Sensors take over rote observation and monitoring tasks from employee |
Employee is freed to focus on higher skilled activities |
| Task Identity |
More insight into the big picture leads to higher satisfaction |
Sensors collect data and displays show process-wide key performance indicators |
Employee is able to feel connected beyond their individual contribution to the entire process |
| Task Significance |
A positive impact on the company or others leads to higher satisfaction |
Sensors collect data and displays show live financial key performance indicators |
Employee is able to feel connected to the big picture and see how they impact the business |
| Autonomy |
More personal decision making leads to higher satisfaction |
Historic data and trend lines show workers the information they need |
Employee is able to make informed decisions |
For a concrete example we can look to manufacturing, say a brewery, and see how this framework is applied. Breweries are essentially batch processing facilities with more or less the same process for each beer, requiring some key ingredient, temperature, timing, and packaging changes from product to product. Temperature and timing are great candidates for IoT monitoring via sensors, while ingredients and packaging are key human value-adds due to the importance of tailoring recipes and the price of raw materials such as bottles.
Using temperature as the first example, without automated temperature monitoring the brewers must check the temperatures, say in the mash tun, at small time intervals and log the temperature on a clipboard. The clipboard is mainly there to make sure they do it, only being consulted if something goes wrong. So, every five minutes the brewer’s stopwatch goes off, they walk over, read the temperature, write it down, then go back to whatever they were doing before. With this much context switching they can’t really focus on anything that well, and reading a temperature gauge doesn’t provide much existential value.
With automated IoT temperature monitoring, thermocouples located in the mash tun log temperature data and feed it to the cloud. If the temperature goes outside of a predetermined range range a notification is sent to the brewer. The brewers are freed to focus on other parts of the process and are happy that they don’t have to be chained to the clipboard. The logged temperatures are displayed on a screen, showing them overall trends and giving them the autonomy to make informed changes to the system.
For the next example, let’s look at the fermentation process, one where you are never quite sure how long it will take. Without automation the beer must be examined periodically over the course of a few days to see if the yeast has finished its work. It is fine if the beer sits too long in the fermentation tank, but it will hold up the next batch, slowing down the overall throughput of the plant. The brewer checks every 30 minutes, but not during the lunch hour, or when they are in a meeting, etc. Eventually fermentation will turn out to be finished and they will move the beer on to the next stage. This isn’t as onerous as the temperature example, but the brewer isn’t engaged to maximize throughput.
With automated IoT monitoring enough data has been collected, sent to the cloud, and analyzed through machine learning to determine that the time to complete fermentation is a function of temperature and batch volume throughout the entire brewing process process. This allows a much better prediction window has been determined, down from days to hours. The brewer now checks frequently during this window, knowing that they have a good chance of catching fermentation right as it finishes. The throughput of the plant is displayed centrally, and the brewers know that their contribution is keeping the plant humming.
These applications play out to any worker in any industry, from energy, to transportation, to retail. Employee retention and satisfaction are not just altruistic goals, but significantly impact the bottom line. Hiring, onboarding, gaining (and losing) institutional knowledge all have high costs. By letting your employees focus on what humans are good at, and by offloading to sensors the tedious parts of their jobs, you uplevel your entire business.