Miller Magazine Issue: 124 April 2020

79 ARTICLE MILLER / APRIL 2020 go beyond graphical representation of data streams. The matching of data patterns and combinations, the ability to compare them automatically with models and well-eva- luated scientific knowledge gives us much better insigh- ts, the ability to predict and generate automated signals, which in the case of an interconnected facility enable op- timal automated control (e.g. optimal grain aeration). Layer 3: Middleware Layer A large portion of what makes IoT paradigm so spe- cial for today’s operations happens at this layer. “IoT” is a very popular term, and, many vendors like to use it a lot, even for solutions that are not really IoT. At the same time customers are more familiarized with stan- dalone monitoring systems that can be installed and report some data or, in the best case, create some grap- hs and warnings based on simple empirical rules, and, a smartphone app. However, this is nothing more than a traditional wireless sensor network (WSN) for monito- ring, not IoT. It is a fact that vendors do not want to undertake the task of building users’ trust towards a new technology or do not want to invest on a real IoT solution yet. At the same time customers to some extend refuse to get out of their comfort zone and famili- arize with something new. This situation suits everybody but at the same time does not help to introduce actual IoT solutions to elevators management, pre- venting all the advantages that those solutions bring together. Two are the most important features of the IoT. One, it is literally an “internet of thin- gs” so much of the information produced in the Layer 1 (sen- sors/actuators) and transferred through Layer 2 (networking) is actually consumed by things and not humans like the inter- net as we know it. The things can be devices like other sen- sors or actuators. Two, IoT hea- vily relies on cloud computing. Super-computers that are che- ap to use as much as you need them but too expensive to maintain in-house, have enough power to run real time analysis on the available data, do intensive computations and run complex algorithms in order to produce predic- tions, assumptions, warnings and optimal control signals. Cloud infrastructure offers access from any device, which is a flexibility valuable for people that must work long hours outside their office, such as elevator managers, gra- in operations managers, grain traders and millers. Grain conditioning which has so far been conducted using fixed set points and rules or even combination of conditions and slightly more complicated logic. The same Figure 2. As it is clearly seen, a large portion of tasks in a real IoT solution take place on the cloud. Picture 3. Centaur wireless gateway and smart sensors.