# Predictive Optimization for Supply Chain

### "There's a way to do it better - find it." Thomas A. Edison

Since 2008, we do our best to deliver the most accurate forecasts that technology can produce. Our technology is in constant evolution to reflect the latest discoveries in mathematics and computer science.

## 6 GENERATIONS OF FORECASTING

Over the last decade, data related technologies have evolved at a crazy pace. Companies went from using technologies that were fundamentally based on mathematics, which had not changed that much since the 18th century, to Big Data oriented technology powered by Machine Learning and Deep Learning. Lokad has been focused on keeping ahead of things and bringing the best science can provide to supply chain optimization.

Take a trip down memory lane and discover the different generations of our forecasting technology.

## THE RIGHT MIX OF INGREDIENTS

### A recipe for success

Lokad's technology is not about leveraging one (or several) magical statistical model. It's a combination of ingredients working together to create the proper alchemy. In our early years, we realized pretty fast how big the gap was between pure mathematical modeling and the reality of supply chains.

What worked wonders in theory was utterly inefficient when applied to real businesses: the data was unclean, not deep enough, too sparse, the sheer volume of references or entries in the sales history for some businesses made entire classes of models extremely hard to use, and then the constraints of the supply chain itself made it so that improving the classical accuracy metrics of the forecasts actually degraded the business' performance.

Lokad had to come up with the proper technological answers to all of these issues and to drastically change its view on forecasting and supply chain optimization.

### Correlations

##### with Deep Learning
When looking at a single product at a time, there is simply not enough data to produce an accurate statistical forecast. Indeed, on most consumer markets, the lifecycle of a product is less than 4 years, which means that, on average, most products don't even have 2 years of history available - that is, the minimal depth to perform a reliable seasonality analysis when looking at a single time-series. We address the problem through statistical correlations: the information obtained on one product helps to refine the forecast of another product. For example, Lokad autodetects the applicable seasonality for a product even if the product has only been sold for 3 months. While no seasonality can be observed with only 3 months of data, if older, longer-lived products are present in the history, then the seasonality can be extracted there and applied to newer products.

### Computing Power

##### through Cloud Computing and GPUs
While leveraging correlations within the historical data vastly improves the accuracy, it also increases the amount of computations to be performed. For example, to correlate 1,000 products looking at all possible pairs, there are a bit less than 1,000,000 combinations. Worse, many companies have a lot more than 1,000 products. By leveraging cloud computing and Graphics Processing Units (GPUs), when clients push their data to us, we allocate the machines just when we need them; then, less than 60min later on, we return the results while we deallocate the machines accordingly. Since the cloud we use (Microsoft Azure) is charging us by the minute, we only consume the capacity that we really need. As no company needs to forecast more than once per day, this strategy cuts hardware cost by more than than 24x compared to traditional approaches.