2014 ERP Software Trends: Big Data
Big data is taking over in retail businesses. The power of big data is as a predictive tool. What color shirt will I purchase tomorrow when I arrive at the store? How can we harness some of that predictive power in our ERP system to help us better run our business? We predict a lot more than retail sales and if our predictions improve, our business processes will improve too. McKinsey says that manufacturers produce two exabytes of data annually. It is time to harvest some of that through ERP big data processes.
Let’s look at an example from the manufacturing world. The routings we use to plan our plant loads, schedule our jobs and our people are based on a somewhat static prediction. They are built on production rates that might be engineered standards or they might be based on our history. Either way, they are a single value representing a rate of production at a work center or an operation. In real life though, production will move at varying rates depending on the person, the equipment, even the weather. What if we could predict more accurately when an operation would be complete through utilizing ERP big data tools? In many cases, we can but most of us aren’t trying.
The New Mindset
Modern production equipment often collects and stores data. That machinery usually can be connected to the internet and the data used as it is collected.
Semiconductor wafers are diced, or cut into the individual devices that control our lives. The saw used is an expensive, high-tech tool with many possible controls available. We should have a standard setting for the blade, the blade RPM and height, the feed rate, the temperature and Ph of the wash solution. Actual throughput is a function of all these plus the individual person operating the saw. During the dicing operation one operator might react to real-time quality differences compared to another and both can be within our operating tolerances.
The data for this capability is available right now. All we need is the development of ERP big data tools to use it and the new mindset to change to a variable schedule.
Bobbi sees some chipping at the edges of the kerf and decides to reduce the feed rate. The edges are better so she runs a little slower. This new rate could be extrapolated and the schedule for following operations could be adjusted based on the revised prediction. The next operation is plating. The schedule says to work on the job when Bobbi is finished. Seeing she will be late, could another lower-priority job be squeezed in the available time and Bobbi’s job still delivered on time?
The data for this capability is available right now. All we need is the development of ERP big data tools to use it and the new mindset to change to a variable schedule. In addition to predictions, the data could be used to feed back to Bobbi as an alert that the reduced feed rate will make her job behind schedule and she might try another method to reduce her chipping.
We won’t get all the way in 2014 but we will start connecting and using the machine data we already have to improve our ERP systems.
Three new technologies your distribution ERP must adapt to
Supply chains are rapidly evolving, so make sure your distribution ERP can adapt to these develop...
Six steps to include in your ERP migration plan
How to construct the smoothest ERP migration plan possible
Five key ERP features for e-commerce businesses
Key ERP features for e-commerce companies to look for when selecting new software