In the past few decades, manufacturers have been able to reduce waste in their production processes and improve product quality and yields by implementing lean and Six Sigma initiatives. However, in industries such as pharmaceuticals and chemicals variability is still a major factor influencing ideal production. Given the complexity of production activities that influence yield in these industries, manufacturers need a more focused approach to improving operations. Advanced analytics provides just this.
In manufacturing, operations managers can use advanced analytics to take a deep dive into process data to identify patterns and relationships among process to improve yields. Many manufacturers now have an abundance of real-time data and the capability to conduct such statistical models. They are taking data sets, linking and aggregating them, and analyzing them to reveal insights.
For example, the production of bio-pharmaceuticals, are manufactured using live, genetically engineered cells. Production teams often monitor over 200 variables to ensure the quality of the the substances being made. Two batches of a particular substance, produced identically, can still have a variation in yield between 50 and 100 percent. This unexplained variability can create issues with capacity and quality.
Advanced analytics have significantly increased the yield in vaccine production while incurring no additional capital expenditures. The process was segmented based on the results of advanced analytics. Various forms of statistical analysis determined interdependencies among the different process parameters and their impact on yield. The manufacturer made targeted process changes to increase its vaccine yield by more than 50 percent and saved the company between $5 million and $10 million annually for a single substance.
Discovering unexpected insights
Even within manufacturing operations that are considered best in class, the use of advanced analytics may reveal further opportunities to increase yield. Several unexpected insights emerged when a company used neural-network techniques (a form of advanced analytics) to simulate the impact of different production inputs on yield. The analysis revealed many unseen sensitivities. By resetting the production parameters accordingly, the chemical company was able to reduce its waste of raw materials by 20 percent and its energy costs by around 15 percent.
Earning more with big data
The first step for manufacturers that want to implement advanced analytics is to consider how much data is available. Many companies collect mountains of process data but use them only for tracking purposes. For these companies, the challenge is to find the experts that can use this data for optimization purposes. Some companies have too little data to be statistically meaningful. The challenge for leaders at these companies will be making a long-term effort to invest in systems to collect more data.
The big data era has only just begun, and advanced analytics has years before becoming standard operating procedure. It is a vital tool for capitalizing on improved yields, particularly in manufacturing environment