PROBLEM STATEMENT: Brunswick is experiencing unprecedented volume for engines and fully produced boats, but the supply chain continues to be difficult to predict and presents interruption to the manufacturing process or brings fluctuations to inventory availability. Brunswick needs to minimize inventory volume and maximize availability of assets for the production process.
PROJECT GOAL: Reduction in unnecessary inventory and closer alignment of asset needs to the manufacturing process. This includes ensuring the availability of inventory where needed, without unnecessary overstock.
TECHNICAL APPROACH: Leverage customer Machine Learning to analyze the production demand against anticipated assets for production to produce a more accurate inventory report. The SM platform will be integrated with a generic data model and create a SM profile that will be used to describe the inputs fields for the supply chain data, such as warehouse, quantity in hand, location, etc. The SM platform will also receive the AI model that predicts the desired supply chain demand using minimum amount of inputs. For accuracy to improve from the generic model, the user will need to provide industry specific data.
KEY TASKS AND MILESTONES:
- Connected Data Environment – Cloud data environment prepared for data sets
- Machine Ready Datasets – Cloud datasets ready for ML workloads
- Data Model for SM Platform – Could dataset model ready for SM platform
- Forecasting Algorithm – Initial algorithm to start iterative ML process
- Predictive Analytics – Iterations on ML algorithm to provide predictive analytics
- ML Model for SM Platform – Availability of the ML model for the SM platform and project close

losses due to non-production states.
BENEFITS:
- $10 – $15 million in inventory reduction & $5m reduction in non-production losses
- Awareness of production process
- Reference architecture for CESMII partners
Member % Cost Share | CESMII % Cost Share | Duration |
50% | 50% | 3 Months |