There’s no doubt that the tools that enable big data and advanced analytics applications—everything from data management and storage platforms to data visualization tools—have, in recent years, matured dramatically, becoming faster, cheaper, and more powerful.
This has opened up a world of new opportunities for manufacturing organizations to understand and optimize their operations in ways that previously were impossible, by gaining insights from the massive amounts of information generated by plant operations and supply chains and quickly acting on those insights.
But how do manufacturing organizations go about taking advantage of those opportunities to make better decisions faster and create agile work cultures? Is it simply a matter of investing in and deploying advanced big data and analytics technologies? Or is there more to it than that?
Last week, members of the Manufacturing Leadership Council visited and toured the Guadalajara, Jalisco, Mexico, campus of IBM, where thousands of team members build high-end storage, power supplies, information appliances, and other products and where the company is in the midst of just such a transition to an agile, data-drive decision-making culture.
IBM leaders in Guadalajara—led by the ML Council’s host Ron Castro, vice president of operations and supply chain execution—emphasized that truly embracing analytics and data-driven decision-making involves much more than simply implementing tools. It is a journey—much like the embrace of lean thinking for continuous improvement—and it requires cultural change, support from top leaders, and the education and engagement of contributors up and down the organization.
IBM is ahead of most manufacturers in making this transition. The company, said Castro and his team, has already implemented a wide range of big data and analytics platforms and systems that help the company orchestrate its very complex supply chain at the individual plant and enterprise levels. Those solutions include descriptive, predictive, and prescriptive systems that help IBM identify and resolve supply, demand, and inventory challenges sometimes even before they blossom into big problems. And IBM has begun to define a new generation of what it calls cognitive applications that not only identify emerging problems and recommend solutions but also actively learn from patterns in structured and unstructured data.
One example is what IBM calls its Transparent Supply Chain initiative. TSC pulls together massive amounts of data from IBM’s systems of record– manufacturing, procurement, fulfillment, logistics and other operational management systems—normalizes it, and uses it as the basis for a wide range of analytics applications and reports that give IBM unprecedented, real-time visibility into supply, demand, and inventory trends and where problems may be brewing. TSC includes an alerting component that uses business rules and system learning capabilities to send targeted information to a user or a group of users via alerts that can be accessed from a watch, a smart phone, or computer via email, text or text to voice.
TSC, which is a 2016 Manufacturing Leadership Award winning project, has spawned applications and use cases across IBM’s manufacturing and supply chain operations and has resulted in huge savings and improved customer service.
IBM is also rolling out a number of cognitive analytics applications that will, for example, use structured and unstructured data, coupled with artificial intelligence to predict and proactively resolve quality and supply chain risk problems.
IBM officials on Castro’s team, however, stressed that cultural and leadership change is every bit as important as technology when it comes to embracing big data and analytics and driving to data-driven decision-making. First, they said, manufacturing leaders need to change the way they make decisions, relying less on experience and intuition and putting more emphasis on empirical data and analytics.
Leaders also need to help contributors up and down the organization not only understand the importance of data-based decision-making but also encourage them to feel empowered to propose and create analytics applications that can drive improvement down to the line or work group level. Workers in IBM’s storage manufacturing operation, for example, created an analytics app that provides demand and inventory visibility during the last two weeks of very quarter when the Guadalajara plant makes a high percentage of its shipments.
But empowering workers to envision and implement such solutions also means leaders must be tolerant of the types of miscues that can accompany innovation, Castro’s team emphasized.
Manufacturers pursuing agile, data-driven decision-making also need to be prepared to enable much higher levels of cross-functional collaboration. Most of the advanced analytics systems deployed by IBM—including TSC—allow different functional parts of the business responsible for assessing demand, placing orders, and procuring materials, for example, to collaborate much more closely and in real time. But that requires the kind of cross-functional processes and trust that, in many organizations, haven’t existed.
It’s also important to realize, said Castro’s team, that analytics don’t replace other types of operational improvement initiatives such as lean and Six Sigma. Analytics complements those continuous efforts while presenting an opportunity for manufacturers to significantly accelerate productivity gains. But only if they can successfully engineer the needed cultural changes.