The encyclopedia · Engineering & Operations · Operational decision · 2003–2004
Nu-kote ran its supply chain on spreadsheet LPs, saving about $1M a year
Spreadsheet linear programs with 5,000–9,700 variables planned Nu-kote's shipments, cutting annual costs about $1M and transit time two days.
Nu-kote International
The solution
Nu-kote International manufactures ink-jet, laser and toner cartridges, ribbons and thermal fax supplies, shipping finished goods across a multi-node network that had grown under ad-hoc distance policies.
Researchers built spreadsheet linear-programming models — 5,000–9,700 variables and 2,452 constraints — to plan shipments between vendors, plants, warehouses and customers, and to test different warehouse configurations.
The models identified improved shipment plans worth about $1 million in annual cost savings and a two-day reduction in customer transit time; Nu-kote saved $425,000 from the first insight the model provided.
Why it worked
- Spreadsheet LPs made the model transparent and usable by existing staff.
- Scenario runs let Nu-kote test warehouse configurations before committing.
- Even the first insight from the model paid $425,000.
- An inherently nonlinear problem was attacked with linear models pragmatically.
What can be applied
Before buying enterprise software, check whether the decision fits in a spreadsheet model; cheap transparent tools get used, and even partial adoption captures most of the value.
Aftermath
Nu-kote used the LP versions to evaluate supply-chain configurations, banking $425,000 immediately with about $1 million in annual savings projected; the case was published in Interfaces (2004).
Sources
- Nu-kote's Spreadsheet Linear-Programming Models for Optimizing Transportation
- Nu-kote's Spreadsheet Linear-Programming Models for Optimizing Transportation
spotted an error? The archive wants to know.