Creating visibility with the cloud

Most manufacturers have made great strides in optimizing operations. Many have even standardized multiple facilities with the same on-premise software to create consistency across the enterprise. However, no matter how many procedures are put in place, each plant has its own idiosyncrasies, making it nearly impossible for systems to communicate with each other. This miscommunication means data cannot be aggregated or summarized across the enterprise, limiting visibility across operations.

Consider that an automotive component supplier might make front-axle half shafts in 20 different facilities. During the manufacturing process, operators must measure various part features to ensure the quality of each piece. If the measurements don’t meet specifications, operators make corrective actions or scrap the part, resulting in wasted time, material and money.

To better manage this process, management might want to know, for example, how much scrap has been generated from out-of-spec ID features across all 20 plants. This seems like a simple request, especially if every plant has standardized on the same software. Here’s the problem: Each plant likely has different naming conventions for the same feature. Plant A may use “inside diameter,” Plant B calls it “diameter – inner,” Plant C denotes it as “ID” and so on.

Something as seemingly simple as displaying waste created from out-of-spec IDs across the enterprise is actually quite difficult. This is because every on-premise deployment is different. In addition to different feature-naming conventions, product codes (although identical from one plant to the next) might be named differently. These differences make it challenging to aggregate data.

Consistent naming conventions create an environment in which analysis of data from multiple plants or sites becomes simple and meaningful.

Furthermore, if data is housed on-site at each plant, aggregated data must be compiled from 20 different databases and placed in a spreadsheet for management to analyze scrap results. Simply put, it’s a mess.