United StatesConstruction and engineering2025 – ongoingClayco
Machine learning for cash-flow forecasting
Clayco cut forecast preparation time by 76%, according to Dataiku
Problem
Clayco estimated cash on hand with spreadsheet formulas that lacked validation and monitoring. Inaccurate forecasts could leave it borrowing for project costs or holding excess cash.
Approach
Its data science team used Dataiku to build an XGBoost model that predicted collection times for open invoices. A custom survival model flagged atypical cases for review. The production workflow automated scoring and retraining, and sent results to accounting dashboards.
Results
Takeaway
The reported gains concerned forecast preparation time and frequency. Dataiku did not provide a numerical measure of forecast accuracy.
Categories
- Industry
- Construction and engineering
- Business function
- Finance, risk and legal
- Innovation type
- Process innovation
- AI technology
- Predictive analytics and ML
- Adoption
- Off-the-shelf tool or platform
- Evidence
- Vendor case study
Sources
Source summaries
Dataiku described Clayco's production forecasting models and accounting dashboards. It reported a 76% reduction in preparation time and a 4x increase in forecast frequency.
Full details are in the original sources.