Analytics Grid


"Retail focused, smart algorithms, high performance."

Array of demand forecasting models

- Multivariate time series models

- Dynamic regression models

- Bayesian models

- Neural Networks

Efficiently manage retail's large scale problems and dimensions

- Very large number of price and non-price variables

- Multilevel distribution channels, store and product hierarchies

- Large set of interdependent and potentially conflicting price positioning rules and performance targets

Comprehensive range of retail specific models

- Demand Forecasting: Daily, weekly, monthly, ...

- Optimising: Prices, promotions, assortments, markdowns, replenishement schedules, safety stocks

- Association Rules learning: Market basket analysis, customers mission profiling

- Clustering & Segmentation: Stores, customers, products

Horizontal and vertical scaling

- Algorithms designed to run in parallel

- Easily implemented on cost-effective public or private cloud based computer services

Data cleaning and analysis

- Completeness, accuracy, relevance

- Distribution analysis, noise evaluation

- Outliers identification and correction