Using Regression to Predict Dyeing Recipe Cost Before the Batch Runs
A simple multiple regression on shade depth, fabric type and machine liquor ratio predicts chemical cost per kilogram within about 6 per cent, which is close enough to quote from.
Business Analytics
Turning operational data into forecasts, dashboards and decisions that hold up in a review meeting.
5 published articles across 5 sub-categories
KPI & DashboardsBreaking
Designing a Factory KPI Dashboard That People Actually Use
Most factory dashboards are abandoned within a quarter. The failure is almost never technical — it is metric definition, refresh timing and the absence of a decision attached to each number.
From Spreadsheet to Forecast: Exponential Smoothing for Demand Planning
You do not need machine learning to forecast most industrial demand. You need the right smoothing model, an honest error metric, and a rule for when to override it.
Statistical Process Control for People Who Do Not Like Statistics
Control charts fail in most factories because the wrong chart is used, limits are recalculated too often, and everyone treats them as specification limits. Here is the version that works.
Choosing an Analytics Stack for a Mid-Sized Manufacturer
A practical decision framework for factories with an ERP, several disconnected machine systems, and a finance team that lives in Excel. Spoiler: the answer is rarely the most impressive tool.