Moreover, today’s global economy and the increasing digitization of information have further increased the complexity of the decision, even as they open up substantial growth opportunities. Layer in the fact that many industrial manufacturers are facing uncertainty in understanding cyclicality, volatility, and customer order patterns, and you can understand the heightened level of risk involved in the decision-making process.
Within this environment, industrial manufacturers understand they need to develop smarter products that can serve as critical solutions within complex, high-stakes manufacturing supply chains. To drive success and integrate their operations in today’s marketplace, they’re harnessing new technologies, from artificial intelligence (AI) and advanced analytics to robotics and additive manufacturing. The rapid adoption of these advanced technologies is leading to disruption in production processes, cost management, order fulfillment, product life cycles, quality control, hiring and training, among other areas.
Are We There Yet?
While advanced technologies are readily being applied within the four walls of manufacturing sites and across supply chains, we’re still not seeing industrial manufacturers leverage advanced technologies to aid in determining network design and site selection. For instance, AI is being embraced in the production environment and in supply chain optimization to capture information about factory performance and predict supply chain disruptions. Yet, there are no packaged AI solutions on the market for optimizing the manufacturing location decision and footprint design.
Companies tend to conduct their own analysis or rely on consulting firms to evaluate the factors important in a site selection decision. Some of this information can be leveraged when planning a new facility or site, but not necessarily when evaluating future site locations. Like IdealSpot, SiteZeus and LLamasoft are bringing together workforce data, traffic analysis, spending information, real estate availability, and machine learning modeling to deliver effective site selection recommendations. Through these products, we’re seeing the emergence of advanced technologies being brought into the site analysis process, but we’re still in the early stages.
Looking at the current model, the process typically includes an analysis of multiple elements spanning operating costs, speed to market, proximity to suppliers, access to talent, supply chain continuity, taxes, competitive responses, and access to technology, among other considerations. These variables, along with considerations on regional macroeconomics, policies, and regulations are included in a robust five-step approach aimed at thoroughly assessing and optimizing complex global manufacturing networks. These steps typically include:
- Understanding of the current situation and underlying economics
- Definition of a short list of scenarios
- Model design and simulation of scenarios
- Synthesis and business case development
- Implementation
AI and machine learning, among other advanced technologies, are being increasingly utilized by industrial manufacturers to improve workflows and supply chain economics, as well as to develop more solutions-oriented products. Integrating Advanced Technologies Into the Process
As advanced technologies are integrated into the site selection process, the future modeling approach should gradually change, allowing for increased use of data and accelerating decisions, thereby enabling increased benefits. For instance, AI could act as an aggregator of “big data” and enable management teams to make more informed decisions about site selection beyond what a typical analysis might determine. Layering geospatial information with supply/demand patterns, cost of distribution, site vulnerability to acts of nature (e.g., earthquakes or flooding), and population densities can lead to new insights on site selection.
Machine learning is a sub-field of AI that uses large amounts of data to discover important patterns. Training the various data inputs is key to making advancements in machine learning, and effective modeling has to be developed to extract the most value from the data on hand. The data exists, but it has not yet been leveraged holistically into a technology-driven platform that can be consistently relied upon to make decisions.
In the case of manufacturing site selection, training data might include current manufacturing footprint inputs spanning locations, square footage, costs, workforce, taxes, and throughput. Economic development agency data would need to include up-to-date information on available properties, workforce, utilities, state and local taxes, and available incentives. Finally, competitive intelligence data might include what other companies in the region and competitive space are doing.
AI and machine learning, among other advanced technologies, are being increasingly utilized by industrial manufacturers to improve workflows and supply chain economics, as well as to develop more solutions-oriented products. However, these advances are not yet being leveraged through any publicly available packaged solutions focused on site selection and design. There remains a clear opportunity for a homegrown system or other third-party solution to serve this need. Given the pace of technological innovation in the sector, we believe it’s only a matter of time before these advancements are harnessed to aid in this increasingly complex decision-making process.