Insights
Articles and analysis from the Lean Path team on demand forecasting, inventory strategy, and the supply chain challenges facing automotive parts suppliers in Japan.
Classical safety stock calculations assume stable lead times and predictable demand. When your customer changes their assembly schedule mid-quarter, those formulas stop protecting you.
Read articleTier-1 suppliers see the schedule. Tier-2 and Tier-3 see only the order. Closing that gap requires translating assembly plant pull signals into forward demand at each upstream tier.
Read articleBattery-electric and hybrid platforms share fewer parts with combustion lines than most planners assumed. The demand mix is shifting, and historical order patterns are losing their predictive power.
Read articleMost supplier demand plans are built from the last 12 months of shipments plus a safety margin. They ignore the assembly plant pull schedule, known model changeovers, and seasonal shutdowns already on the calendar.
Read articleOverstock is not just tied-up cash. It crowds out production capacity, inflates carrying costs, and hides forecast problems until the next model change forces a write-down.
Read articlePure just-in-time works when supply chains are predictable. Just-in-case buffers work when they are not. The question is how large those buffers need to be and when to activate them.
Read articleAI forecasting is not a replacement for planning experience. It is a tool for translating more signals into more accurate orders. Here is what mid-size suppliers need to get started.
Read articleA sudden production surge at the assembly plant should not require emergency orders two tiers up. When it does, it means demand signals are not reaching suppliers in time to act.
Read articleMost auto parts suppliers do not have machine learning engineers. Forecast accuracy at scale does not require them. What it does require is the right data and a model trained on your supply chain.
Read articleBetter demand visibility does not require suppliers to rebuild their relationships with assembly plants. It requires better use of the data those plants already generate every day.
Read articlePull-based production systems generate a specific kind of demand signal. AI models that understand that signal can forecast more accurately than models trained on generic demand patterns.
Read articleThe planning tools used by Tier-1 suppliers were not designed for Tier-2 and Tier-3 realities. A new generation of demand forecasting needs to start from the suppliers that conventional tools have always underserved.
Read article
