In the era of Intralogistics 4.0, companies will turn to “big data” analysis to find ways to fix problems and improve productivity in their distribution centers (DCs.) Big data analysis is the process of examining large data sets to discover hidden patterns, unknown correlations, and obscure connections. By identifying relationships in data, companies can determine the factor or mix of factors that are inhibiting performance or causing a problem.
DCs are ripe for big data analysis because there is a tremendous amount of transactional data that can be exploited to identify cause and effect relationships and ultimately improve operational efficiency. In fact, some warehouse execution software systems overseeing distribution and fulfillment operations compile huge logs of transactional data that provide an audit trail of all warehouse activity and inputs, including receiving, material movement, handling, picking, packing, and shipping all by order, item, worker, and length of time. Tools such as data mining and statistical analysis can be applied to warehouse transaction logs to discover patterns, correlations, and connections that could enhance productivity.
At MSI Automate, we’re already using big data analysis to help our clients gain actionable insights to increase DC throughput and productivity. Since MSI Automate’s warehouse execution software already logs huge amounts of transactional data, our Intralogistics Science team is able to use the data to perform a “trending analysis,” which is the discovery of commonality in a data pattern that denotes when a combination of factors produces a certain outcome. To perform this type of analysis, the team uses “R” programming, a special computer language for statistical analysis and data visualization used by data scientists. R Programming was developed specifically to mine through data logs and find correlations between factors that are influencing or causing a specific outcome. The results of the trending analysis can then be applied to determine the root cause of a problem in the DC, or, simply to find more optimal ways to perform DC operations.
But big data analysis is just the first step in solving a problem through the use of Intralogistics Science. Once the data analysis has pinpointed the root cause of a problem, the team at MSI Automate then uses simulation to test various hypothetical remedies. Some remedies may incorporate a design change or process change, or both. A computer model is then used to emulate how the proposed solution would perform in the DC to validate its effectiveness. By providing a emulation, MSI Automate enables its client to ascertain whether the solution has merit before making any operational changes or incurring any expenses associated with the changes.
Because the growth of ecommerce in combination with rising consumer expectation for same-day delivery will pressure DCs to speed up order fulfillment in the era of Intralogistics 4.0, big data analysis promises enormous potential for companies to gain the insights needed to meet that challenge and gain a competitive advantage over their rivals in the marketplace.
Walter High is VP Marketing at MSI Automate, where he has worked since 2012.