Forecasting + Inventory Optimization: JCI Parts

PROBLEM / CHALLENGE

Johnson Controls Parts NA business unit was experiencing a combination of increased inventory and backlog. They had too much of the wrong inventory and not enough of the right inventory. This problem became exacerbated during the pandemic as a result of the supply chain shortages and increased and highly variable lead times. They were utilizing an excel based forecasting model to manage their inventory decisions.

IMPLEMENTATION DETAILS: FORECASTING + INVENTORY OPTIMIZATION

• In the assessment of their processes, we identified the limitations of their forecasting and inventory model:
  • Lack of seasonality component
  • Inability to handle variable lead times
  • Limited historical data
  • Challenges with pandemic disruption to forecast
• Developed a forecast using optimization to minimize forecast error across all 50,000 SKUs
• With forecast, target customer service levels, and lead times, we calculated the target reorder point (ROP) and reorder quantity (ROQ) by SKU
• Due to capital limitations on inventory investment, we developed an inventory rebalancing process to support the reduction of the slow-moving inventory and allow for the purchasing of more fast-selling inventory

DIFFICULTIES ENCOUNTERED

• Variability in lead time and demand, as a result of the pandemic, was challenging when calculating reasonable ROPs and ROQs
• Financial limitations on inventory investment, delayed the company’s ability to get to a strong inventory position. A significant amount of work was required to sell down obsolete inventory to make
• room (physically and financially) for the new inventory

RESULTS

• Improved forecast accuracy
• Streamlined processes for procurement team as purchase recommendations from new forecast were more aligned with buyers’ expectations.
• Improved ability to service customers with in-stock product