We Optimise Fresh

Current forecasting methods are generally manual and based on historical data only. Typical forecasting methods involve manually identifying the same week in a previous year or year(s), which are considered to be similar to the current year, and manually adjusting based on experience and gut feel to account for new information like competitive activity, price movements, promotions and weather.

Simple statistical forecasting often doesn’t provide an improvement over manual methods based on experience.

However, machine learning based methods allow more variables such as broader price data, weather data, public holidays, geographical and socioeconomic data to be incorporated leading to greater accuracy than manual methods.

Fresh is one of the key drivers of growth for retailers. Traditional promotional planning systems are not suited to the continous supply and demand management that is needed for fresh, and are based on demand only. This makes it hard to make optimal decisions accounting for both supply and demand, leading to losses for growers and unnecessary discounting.

Fresh specific promotional planning allows greater flexibility in managing promotions and incorporating the latest supply and demand information.

Optimising inventory means determining the exact quantity of stock for a product which is most likely to meet demand and therefore minimise losses due to overstocking and waste, while also minimising losses due to understocking and lost sales.

As demand is always variable and can never be predicted perfectly, optimisation aims to weigh up understocking costs like lost sales against overstocking costs like waste. Optimisation is always based on a demand forecast which can be manually created, or the result of a statistical or machine learning model.