The scope and quality of data that enterprises collect and access in their supply chains, and how they leverage that data to optimize their supply chains, differs significantly from one enterprise to the other. On a recent day, I had conversations with a few OEMs and reviewed one particular inventory report that were all insightful on this very topic.
In one of the conversations, the director of a supply chain discussed her team’s ability to view component inventory across multiple contract manufacturing sites every morning, and how they took steps to re-balance it in accordance with manufacturing plans.
Just a few hours prior to that, in a conversation with another OEM, the supply chain manager lamented that inventory re-balancing tasks were reactive for his team, and the inventory reports he received from his contract manufacturing partners were about a week old by the time his team member had scrubbed them and was ready to analyze them.
supply chain systems and data warehouses to understand trends.
Later that evening, I reviewed a transaction where the planning system had generated and executed an automatic inter-site purchase order to align component inventory with site manufacturing plans. In each case, OEMs had access to similar information, though with different latency.
At one extreme, significant effort and time was being investeded in dealing with the consequences of inventory imbalances due to fluctuations in demand plans. At the other end, using sets of tools and techniques, data from multiple sources was being analyzed in close to real-time, and decisions were being made. In this case, technology played a critical role in assimilating large and complex data sets from upstream and downstream members of the value chain, who analyzed it and modeled it to optimize the supply chain.
These cases outline the wide range of analytics capabilities that enterprises have. Standard and ad-hoc reports are used primarily to answer questions such as, “What happened?”
More evolved solutions and capabilities leverage OLAP (Online Analytical Processing) tools to analyze data in multiple dimensions and answer questions like, “Why did it happen?”
At the top end of analytics capabilities (also called advanced analytics) is predictive modeling, which allows enterprises to look for trends or patterns in data from across the value chain, and answer forward-looking questions like, “What would happen if…?”
These capabilities result from leveraging data that resides in multiple systems within the enterprise, as well as data from partner or external systems, and combining it with the analysis of vast realms of data in near real-time. Specifically, three key technological forces continue to impact and drive wider adoption of Advanced Analytics — cloud computing (supports collection and storage of data from multiple systems and partners), in-memory computing (supports fast data processing), and improved modeling tools (easier to implement and better user experience).
Application of Advanced Analytics is not constrained to a specific process in the supply chain. Operational benefits that focus on cost optimization, improved productivity, and higher customer satisfaction can be derived for any supply chain process including procurement, supplier management, order management, demand management, transportation, and returns management.
Yet, as the conversation I shared earlier demonstrates, adoption and deployment rates for analytics tools in the supply chain processes of the hi-tech industry could be better. A host of reasons could be attributed to slow deployment, though the following two stand out:
The first reason relates to the quality and timeliness of data in extended global supply chains. Enterprises with limited access to upstream and downstream partner data find it difficult to justify the addition of Advanced Analytics to their roadmap. Even when fair, quality data is available, some enterprises are unsure of their capability to integrate their ERP, data warehouses, and supply chain management systems to the Analytics tools.
The latter reason relates to the shortage of talent with deep analytical skills, as highlighted in the “Big data: The next frontier for innovation, competition, and productivity” report (page 103) from McKinsey Global Institute. The talent required to deliver value from Advanced Analytics must have relevant analysis and statistics skills, but must also have the necessary business and vertical acumen to interpret and communicate the output from modeling tools.
In the high-tech industry, distributors like Avnet have traditionally leveraged structured data (including partner data) residing in ERP, supply chain systems, and data warehouses, to understand trends and share the resulting intelligence with customers and suppliers. Distributors are now leveraging in-memory computing technology and new generation of visualization tools to create value for their business partners.
I invite you to share your insight and experiences related to application of Advanced Analytics in the high-tech industry’s supply chains.