Choosing the right forecasting metric is not straightforward. Let’s review the pro and con of RMSE, MAE, MAPE, and Bias. Spoiler: MAPE is the worst. Don’t use it.

  • Bias represents the historical average…


To set up a perfect forecasting process you need to get 4 things right: granularity, temporality, metrics and process.

The 4-Dimensions Forecasting Framework

Demand Forecasting to Support Decision-Making


Everyone is aware of the recent Deep Learning achievements. But Neural Networks have a long history starting 80 years ago.

1940 Prehistory: Artificial Neurons

1950 Artificial Neural Networks


What are the best practices to set up safety stock targets in a supply chain?

Inventory Maturity Level: 5 Stages. Copyright: Nicolas Vandeput


Data Science will allow demand planners to bring their forecast accuracy to unprecedented levels. Learning data science is possible for anyone — but will require time.

Taking Demand Planning Skills to the Next Level

1. Machine Learning is easy to use. You can do it.


In this article, I’ll show you five ways to load data in Python. Achieving a speedup of 3 orders of magnitude.

Source: https://www.hippopx.com/, public domain

Experimental Setup


Operational Decisions based on Demand Forecasts


As a demand planner, you can only focus on a limited forecasting horizon. How long should it be?


In this short, straightforward guide, I’ll show you how to use your GPU to run XGBoost on a windows machine. We’ll speed up training time by 4.

How much faster is a GPU?

from xgboost.sklearn import XGBRegressor
import numpy as np
import time
start = time.time()
XGB = XGBRegressor(tree_method = "hist")
XGB = XGB.fit(np.random.randint(0,100,size=(10000000,30)), np.random.randint(0,10,size=(10000000)))
print(time.time() - start)
start = time.time()
XGB = XGBRegressor(tree_method = "gpu_hist")
XGB = XGB.fit(np.random.randint(0,100,size=(10000000,30)), np.random.randint(0,10,size=(10000000)))
print(time.time() - start)…


Forecasting KPIs such as MAPE, MAE, and RMSE are not suited to assess the accuracy of a product portfolio. Let’s take a look at a few new metrics: MASE, RMSSE, WMASE, and WRMSSE.

Nicolas Vandeput

Consultant, Coach, Trainer, Author: 📙Data Science & Forecasting, 📘Inventory Optimization www.linkedin.com/in/vandeputnicolas

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store