Intelligent Forecasting: 2 hr briefing

Quadbase Systems Inc

Overview and demo of forecasting methods such as Auto ARIMA and LSTM multivariate time series forecasting using Python and Microsoft Azure ML Services.

The importance of accurate forecasting in today’s competitive business world cannot be understated. For a retail business, brick-and-mortar or otherwise, and food service business, inventory management is the key to profit and loss. Too little inventory on hand can result in missing sales opportunities and unhappy customers. Too much in stock can result in unnecessary tie-up of capital, spoilage, and erosion of profitability. A hotel operator would like to get accurate forecast for bookings and occupancy rate so that he/she can lower labor costs by setting up efficient employee schedules that match demand and optimize room rates. Companies such as Apple have demonstrated that supply chain capabilities that include forecasting can be a source of superior revenue growth and market share increase, in addition to lower costs. Revenue is increased and customer satisfaction improved due to minimum stock-outs. Inventory carrying costs, rush order production and expedited shipment costs are minimized.

We invite you to spend a couple of hours with our team of experts in AI, and machine learning to understand your operational issues, pain points and discuss how we can help you utilize AI technologies to improve your forecasting accuracy. As your trusted Microsoft partner, we leverage on, but not limited to, Microsoft AI/ML technologies such as Azure ML services, Azure Databricks, Python, , Microsoft R, Microsoft SQL Server and Power Platform. We will use all relevant data including static and temporal data that can affect outcomes. We will build models with different algorithms such as, Auto Arima, LSTM neural network and regressions. Training results will be compared for selection of the most optimized model. Prediction result in production deployment can be visualized as reports and charts in dashboards.


  • Introduction to Azure Machine Learning, AI and time series data analysis
  • Demo a simple multi variate time series forecasting example
  • Discussion on your forecasting requirements and business issues that you would like to address
  • Discussion of performance metric improvement areas and possible solutions


  • Statement of understanding of your business cases
  • Recommendations on best possible solutions to your business problems
  • Plan for further discussion and/or POC, as required

Please click here to see a demo.