Several years after the global movement of goods came to a screeching halt, this summer brought some good news regarding global supply chains, as the Federal Reserve Bank of New York’s Global Supply Chain Pressure Index remained stable in neutral territory, as has been the case since spring 2017. 2022.
The relative calm of indicators like this suggests that organizations around the world, including those in the electronics industry, are honoring the “never again” vows they made in the wake of the extreme disruptions of the pandemic by taking aggressive steps to protect and strengthen their businesses. their supply chains in the face of geopolitical, climate, labor and other factors that continue to weigh heavily.
Source: Getty Images
Many of these measures aim to make their supply chains more intelligent, automated and data-driven. In a November 2023 survey, Gartner found that nearly two-thirds of supply chain leaders plan to implement or are already in the process of implementing generative AI technology, and that they are spending on average 5.8% of their budget to implement generative AI technology. genAI in their IT in order to improve business processes.
Interest in genAI and other intelligent capabilities to support supply chain management is certainly growing among the electronics manufacturers I work with. They realized that the better they understand and forecast demand for their products, using AI-based precision analysis and forecasting tools, the better they will be able to mitigate the impact of disruptions and to maintain the integrity of their supply chains.
To achieve this, they focus on these five areas:
Liberate and unify
- Free and unify distributed data trapped in disparate locations. Data widely dispersed across a patchwork of unintegrated systems within an organization is the enemy of accurate demand forecasting and an intelligent, efficient business. To produce timely, accurate and relevant demand forecasts, people and the models behind them must have current, comprehensive data available, when they need it, from anywhere in the enterprise. For example, having real-time insight into inventory levels across the growing number of dispersed distribution centers a company may have. Collecting, processing and digesting all this data becomes completely manageable when business systems, data and processes are integrated and cloud-based.
Synthesize multiple internal and external data streams
- Quickly leverage and synthesize multiple internal and external data streams that feed into an enterprise-wide data reservoir. Data, both qualitative and quantitative, structured and unstructured, from traditional as well as less conventional sources, is the lifeblood of superior demand forecasting. Quantitative data typically comes from internal sources and can include highly segmented sales figures, pricing and margin data, customer feedback, marketing campaign data, and web and search analytics. Qualitative data, on the other hand, tends to come more from external sources: news reports, social media platforms, reports on competitor activity, market research, and even weather forecasts. Business AI is particularly adept at using this external data to predict disruptions. Instead of relying primarily on traditional retrospective data, past experience, and instinct to create forecasts, intelligent data management tools can extract nonlinear data from these sources and then synthesize it into a pool of unique data that AI tools can easily access.
AI models
- Use AI-based models to identify correlations, dependencies and patterns between different variables, thereby producing more concise demand forecasts. What happens to all this data once it’s synthesized? This is where the latest wave of AI-driven analytics tools can really improve your organization’s demand forecasting and supply chain planning. These tools can quickly analyze and process deep and complex data sets, spotting relationships and trends that would otherwise likely be overlooked, and then integrating these signals into demand and inventory forecasts and plans. Again, it’s about basing decisions more on data and evidence than on instinct and experience. Instead of just knowing that a key factor driving demand forecasting has changed, decision makers understand why that key factor has changed, leading not only to better forecasting but also to better overall supply chain management, thereby avoiding the bullwhip effect that hit electronics supply chains during the pandemic. .
Encourage data sharing
- Increase trust, cooperation and collaboration throughout the supply chain to encourage data sharing and improve visibility. To maximize the value of the information produced by the aforementioned models, it is essential to populate them with new data from relevant upstream and downstream entities throughout the supply chain. In today’s interdependent global supply chains, it is critical for an electronics manufacturer to know well in advance that one of its major suppliers is expecting dangerously low inventory levels for a key component. a product, for example, so that it can turn to other suppliers if necessary. This type of data sharing and collaboration between manufacturers and multiple tiers of suppliers is increasing across various industries (semiconductor, automotive, etc.), enabled and encouraged by vertically connected consortia whose members are willing (and able) to share data in real time. data on supply and demand, without endangering their competitive data. Using data collected across the value chain, business AI can identify certain dynamics that indicate inbound bullwhip effects in a supply chain, so manufacturers can prepare accordingly.
New approaches to better manage
- Explore new, intelligent approaches to better manage the factors that go into the demand forecasting equation. Beyond applying AI and ML tools to gain deeper insights from data to guide their demand forecasts, other emerging approaches can help electronics manufacturers strengthen their demand forecasts. For example, an AI-driven process known as synthetic paneling uses generative AI to create and derive insights from artificial customer personas (instead of human panelists) to guide product development, decisions marketing and demand forecasts. GenAI acts as a moderator and provides panelists’ responses based on its analysis of data from sources like those mentioned above. In terms of cost, speed and depth of analysis, this is a process that could radically change the way companies approach market research and demand forecasting.
Stable supply chains
- It is in times like these, where supply chains appear relatively stable, that electronics companies would be wise to take steps to refine their demand forecasts and overall supply chain planning . Because as we all know, history has a tendency to repeat itself.
————————-
As solution manager for the high-tech industry division at SAP, Dominik Erlebach specializes in supply networks, enterprise networks and business applications of AI.