Learn How IoT Systems Make Real-Time Decisions

on July 28, 2016
IoT System Example

Transforming Real-Time Data from IoT Systems into Real-Time Decisions

One of our partners, Mariner, implements solutions for Internet of Things (IoT) systems for their manufacturing and distribution customers. They collect, aggregate, and enhance data from devices and sensors using Microsoft Azure technologies. They then pass the enriched data into a SMARTS™ decision service.

In this post, we’ll explore how Mariner uses SMARTS™ decision management platform to make real-time decisions.

IoT System Customer Example

Let’s look at an actual customer story. ABT Power Management is an innovative power management company based in North Carolina. ABT has a service offering, GuaranteedPOWER®, to proactively manage the batteries used in lift trucks and other material handling vehicles. When batteries aren’t properly maintained, repairing and replacing them can actually cost more than the vehicles they power. This offering leverages an IoT system solution developed by Mariner.

As the high-level architecture diagram illustrates, sensor data from batteries and chargers is collected and enhanced prior to being passed to a SMARTS™ decision service. SMARTS™ alerts ABT to dispatch field engineers to the site when equipment maintenance or repair is needed.

The knowledge on how to properly maintain the batteries and chargers is based on the manufacturers’ recommendations and the expertise of ABT’s field engineers. The engineers have a deep understanding of the technology and years of real-world experience. Mariner worked with the engineers to translate their knowledge into concrete SMARTS™ decisions that define how to proactively manage and service the batteries and chargers. These decisions are made up of business rules that define when to perform specific maintenance services such as charging, watering, and rotating the batteries at a customer’s site.

Capturing Knowledge from IoT Systems

Let’s look at a simple example. One condition the sensors monitor is the water level in the batteries. When the water level is low, the battery needs water. You could write a simple rule to capture this:

IF battery water level is low
THEN create an alert to water the battery

But in the real world, this rule is not quite as simple as it seems at first glance! Battery sensors are very sensitive instruments and when a battery is being physically moved it could register a low water condition. So the rules need to take this knowledge into account and perhaps detect if the battery water level registers low for two or more consecutive days. Also the rules need to consider additional factors in deciding whether or not to water the battery. For example, how many other maintenance actions are required at a site? How much time does it take and how much does it cost to dispatch a field engineer to the site? And, can the maintenance visit be scheduled so that it coincides with maintenance and repair required by other customers who are in the same geographical area?

All of this knowledge is captured in the decision making rules that analyze the sensor data. In addition, predictive analytics is used to detect patterns in the data that could lead to future equipment failures. Preventative maintenance is scheduled to prevent these failures.

Mariner’s business analysts captured and tested these rules and decisions in the SMARTS™ Analyst Workbench.

Mariner also used SMARTS™ to run simulations using historical data in order to ensure that the decisions (and resulting alerts) were consistent with the recommended actions specified by ABT’s field engineers.

Deploying and Continuously Improving

Once tested and validated, the rules were deployed to the SMARTS™ decision service, where they make proactive maintenance and servicing decisions in real time. Over time, new rules have been added and existing rules have been improved and refined so that more conditions can be automatically detected and acted upon.

In summary, SMARTS™ Decision Manager provides an ideal platform for automating and deploying IoT real-time decisions. SMARTS™ helps organizations make sense of vast amounts of sensor data and translate it into concrete actions.

Learn more about Decision Management and Sparkling Logic’s SMARTS™ Data-Powered Decision Manager

Search Posts by Category


Sparkling Logic Inc. is a Silicon Valley-based company dedicated to helping organizations automate and optimize key decisions in daily business operations and customer interactions in a low-code, no-code environment. Our core product, SMARTS™ Data-Powered Decision Manager, is an all-in-one decision management platform designed for business analysts to quickly automate and continuously optimize complex operational decisions. Learn more by requesting a live demo or free trial today.