The Indusa Blog

Enhance Predictive Analytics in Your ERP with Azure Machine Learning

Posted by Indusa Admin on August 9, 2016 1:09 pm

Share Button

Machine learning is in high demand, as customers are progressively looking for ways to gather insights from their data. It is at the peak of the Gartner Hype Cycle for Emerging Technologies, 2015. It uses computers to run predictive models that learn from the existing data in order to forecast future outcomes, trends, and behaviors. Overall these forecasts make the apps and devices smarter.

Azure Machine Learning (Azure ML) is a powerful cloud-based predictive analytics service that makes it possible to quickly create and deploy predictive models as analytics solutions. Predictive analytics is the underlying technology behind Azure Machine Learning and can be defined as a way to analyze collected or current data for patterns or trends in order to forecast future events and driving the desired outcomes. Azure Machine Learning provides a fully-managed service you can use to deploy your predictive models as ready-to-consume web services, in addition to providing tools to create complete predictive analytics solutions in the cloud.

azure machine learning
azure machine learning 2
More and more ERP users are adopting machine learning to improve the way their businesses run. Manufacturers identify new mechanisms to alert so they can act on KPIs in order to reach optimum productivity. With predictive analytics and demand forecasting using Azure Machine Learning for Dynamics AX ERP, you can make near real-time decisions through a host of devices that access these KPIs. A chain of decisions from machine learning can help answer higher level questions like:

  1. Which is the best distribution center to service an incoming customer order from?
  2. What is the best price for a product given customer and channel?
  3. How best to price a service contract?
  4. Who are our most valuable customers?

The Power BI capabilities in Dynamics AX ERP provide real-time analytics for predicting behaviors with the power of Azure Machine Learning. These analytics could help a retail establishment carry out smart inventory changes or provide product recommendations via dashboards similar to the one shown below.

CFO_Role_Center_with_Power_View_Map
More opportunities to add machine learning- driven recommendations and prediction engines help organizations deliver better performance and decision making in existing business processes like e-commerce, customer service, and demand forecasting. The solution also provides monitoring and trending analytics prioritized by the manufacturer enhancing the speed of business.

Azure Machine Learning enhances productivity
A leading motor manufacturer was looking to replace or enhance monitoring with lower cost and higher productivity. As a Dynamics AX ERP user at the forefront of using technology, they wanted to enhance their business models:

  1. For direct sales on equipment or directly to the end user
  2. For service partner sale
    • Enhance position in distribution channel
    • Penetration through service differentiation
    • Increased end user specification through monitoring

The manufacturer wanted to target unmonitored equipment that was not wired into a process system and had non-process system applications.

With Azure Machine Learning for Dynamics AX, the motor manufacturer was able to increase the productivity of its sensors in terms of initial setup for vibration and temperature. The communication gateway was optimized to receive sensor data using appropriate wireless communication interface and to store and transmit data received from various sensors. This was facilitated with a cloud-based IoT platform and Azure cloud-based infrastructure with SQL and time series database. Dashboards, reports, and notifications along with user management modules for visualization with Azure Machine Learning helped the client with predictive analytics and demand forecasting using the data fetched from Dynamics AX.

With Azure Machine Learning, the manufacturer can create revenue streams using connected capabilities and integrating Machine Health monitoring. The Azure Stream Analytics provides anomaly detection. By surfacing Azure ML within Dynamics AX, the manufacturer has improved service and quality using connected data and cost reduction and best-in quality services through managed after sales support network. Even better, the manufacturer avoids revenue loss due to motor failure and reduced warranty cost, and inventory on spares. Support for performing analytics of motor usage, customer usage pattern, motor failures, and location information resulted in key differentiators. CSsnapshot azure  machine learning

The Road Ahead

Alexander Linden, research vice president at Gartner says, “Today, we struggle to find 10 business applications that do not embed machine learning,” reiterating the fact that digital businesses are increasingly adopting machine learning. The result? With machine learning in ERP, manufacturers can now achieve a host of preferential insights and predict possible troubles in managing lifecycles and optimizing inventory levels. Machine learning is creating the “New Dynamics AX” ERP, embedding Dynamics AX with smart analytics, and allowing manufacturers to identify profitable opportunities with demand forecasting and predictive analytics. Blog-callout azure machine learning

Can machine learning create automated, intelligent processes for your business? Read this ebook to know more about the New Dynamics AX that has been upgraded with new capabilities in Azure Machine Learning and Power BI.

New Call-to-action


Manoj Nair
About the Author – Manoj Nair

Manoj Nair is a seasoned IT professional with over 12 years of experience in Software Consultancy, ERP Implementation, Testing, ISV Development & Client Management. He is a Senior Project Manager at Indusa and is responsible of managing various software programs through cross functional coordination.

connect-on-linkedin

Contributing Writer: Nandita Nityanandam


Topics: ERP