Dematic iQ InSights: Smart analysis of warehouse data

The Dematic iQ InSights project gained momentum over the past year for the simple reason that there is a demand for it. “It was really the customers asking for more: more information, more visibility, more analytics, more metrics,” he explains.

Visibility was right at the top of the wish list. For many warehouse managers, the lack of a comprehensive overview of all processes at their facilities is the biggest problem. But obtaining this overview is very difficult to achieve in massive halls using conventional methods. In some cases, there are catwalks or an office overlooking the hall, but it is difficult to keep an eye on all processes due to the constantly moving multi-shuttles, working platforms and the aisles.

It can take 15 minutes to drive from one end of a warehouse to the other to check where the machinery is reporting an overload, why a vertical conveyor belt has stopped, or what is causing problems for the colleagues in packaging. “When we visited customers, they told us how they don’t have visibility across their distribution center,” says Laura Fraser, who works on Terrenzio’s team in Waterloo.

For Fraser, the benefits of DiQ inSights are obvious: until now, software only provided descriptive analytics about a system. Experts would then combine the data and make decisions based on their experience. Displaying everything on a single screen was nearly impossible since the various programs were not compatible.

Thanks to predictive analytics, the DiQ software makes life much easier for managers and technicians. Data from every part of the system, including multi-shuttles, conveyor belts, sensors and IoT devices is analyzed in parallel and in real time, allowing customers to predict precisely what will happen and where.

“If you can picture a facility measuring one million square feet – or about 18 US football fields – on the size of a tablet, then it is a lot easier to get your head wrapped around,” Terrenzio says. After extensive discussions with customers, the programmers designed the app’s layout in a way that is logical to anyone in the industry. “People have mental models about how something works. In a distribution center, you will find areas for picking, packing, receiving, and shipping. We used that as our inspiration,” Fraser explains.

If an area flashes yellow or red, then users have the option to drill down – zoom in, look at the details, and know immediately where the problem is. The app creates a job for the maintenance crew, tasks a technician with the repair via email or text, monitors the progress and calculates the impact on successive work processes. The operator can concentrate on resolving the technical fault instead of wasting time trying to locate the issue.

And Dematic iQ InSights can do even more, for example identify operational weaknesses in the event of bottlenecks or overcapacity, when the flow of goods is interrupted, or when there is a logistical hitch. “If you need 15 workers to do these picks, and you’re currently short five, that’s going to have an impact,” says Fraser.

DiQ software can calculate individual equipment limits based on volume, priority, weight, or operator role. The warehouse manager receives an email or text message outlining the capacity bottlenecks and can react accordingly.

For Terrenzio, the solution’s greatest achievement is removing the guesswork from the process. “Humans make too many gut decisions,” he says. By using analytical and empirical data, improvement suggestions can be much more specific. For example: ‘It makes more sense to put the picking area in a different part of the hall’ or ‘These particular goods should have packing priority as it works better logistically.’

DiQ InSights can be integrated with existing company systems that are tried and tested, which is particularly attractive for KION customers. “We don’t force the customers into our ecosystem, we can work with other players as well,” Terrenzio says.

Eventually, Dematic wants to achieve a “digital twin”, an exact digital replica on screen of what is happening in real time. Reaching that goal will be the starting point for the next milestone in the software’s development: artificial intelligence (AI) and machine learning.