Recently, traditional manufacturing, wherein raw materials are joined together in a multi-phase, product creation process, has taken a backseat to smart manufacturing. Smart manufacturing builds on the historical conception of manufacturing as a process- and materials-driven construction, adding information technology, flexibility, and computer control.
This article discusses the burgeoning field of smart manufacturing and, more specifically, the increasingly relevant role of big data in helping product development teams across industries to improve processes, control costs, design products, and align production supply with consumer demand. Examining the interplay between smart manufacturing and big data analytics, this article describes the current smart manufacturing and big data environment, explores future developments, and identifies big data’s limitations in smart manufacturing.
Companies increasingly use technology to capture data about their products. Supply chain logistics, tracking customer details, detecting product defects using optical sensors, truck fuel consumption and speed — these are all areas in which manufacturing companies transform collected data into valuable business insights.
However, data is often poorly organized in disjointed databases, wasting time and resources. The current relationship between smart manufacturing and big data analytics suffers from a lack of qualified personnel, inadequate industry and government investment, data privacy concerns, and the practicality of long-term storage of high-quality data. Predictive modeling, while currently in an emergent stage, remains too often superseded by “reactive” modeling; that is, collecting data on processes that have already been executed, rather than making predictions to avoid the re-occurrence of undesirable outcomes. Future developments in smart manufacturing as it relates to big data analytics must correct these weaknesses, and firms must determine if there is a valuable application of this partnership of emerging technologies in their own organizations.
In describing the current state of smart manufacturing and its intersection with big data analytics, we can begin to make inferences about potential future developments. First, companies must learn to organize their data more effectively. Unified database management systems can recover time wasted in searching for information across irrelevant sources. Moreover, firms need to ensure that they have hired the best-qualified technical personnel to handle data. Many undergraduate business degrees severely lack analytics and data science training that is relevant to the marketplace, relying on either an entirely theoretical approach or putting concepts into practice in professionally useless ways.
In addition, some evidence suggests that for smart manufacturing and big data analytics to partner most effectively and serve as an example for future developments, they must be implemented in ways that would serve the public interest to attract government investment. Health care, security/cybersecurity, and renewable energy deserve special emphasis: All are crucial to social and economic development, yet are often inefficient in both cost and delivery.