メインコンテンツに飛ぶ

We'd prefer it if you saw us at our best.

Pega.com is not optimized for Internet Explorer. For the optimal experience, please use:

Close Deprecation Notice
event streaming

Put your data to work with Pega Process AI

Ryan Easley, ログインしてブログを購読する

There is an enormous amount of data being created every day. According to Earthweb, people on the internet generate 25 quintillion bytes daily (Wise, 2022). This figure doesn’t include devices, vehicles, or machines! Many companies have a difficult time keeping up with the volume and velocity of data. In addition to collecting the data, there is also the consideration of what to actually do with the data. How can companies make sense of all this data, and use it to their advantage?

Pega Process AI with Event Strategy Manager (ESM) provides a way to get insights from real-time streaming data and act on it – all through Pega’s low-code platform. There are three components that enable this:

  • Data Flows
  • Event Strategies
  • Decision Strategies

1. Data Flows ingest data, transform it in several ways, and move the data to the right place.

data flow

Data Flows provide a flexible, scalable data pipeline to ingest, transform, and output data to the right place. With Data Flows you can connect to a Kafka topic, perform calculations on the data, filter certain events, start a workflow, and send the event data to Hadoop.

Data Flows can also leverage a feature called Event Strategies, which detect patterns in event data over a window of time. They can filter, aggregate, and identify the events that are important. For example, a mobile provider can use Event Strategy to sift through phone calls for each customer, looking for dropped phone calls over a period of three days. If a customer has more than five dropped calls over those three days, then a response or action on the event is required.

2. Event Strategies detect patterns of data over a window of time and respond to events when these patterns are detected.

event strategy

Data Flows and Event Strategies can be combined with Pega’s Decision Management capabilities to make AI-powered decisions on the data. This is done with Decision Strategies. Decision Strategies combine business rules with Pega’s self-learning Adaptive models or operationalize existing models to help determine the next best action to take in response to an event.

3. Decision Strategies combine business rules with analytics to determine the best course of action to take on the event.

decision strategy

There are many uses for Event Strategy Manager ESM. Two such use cases are predictive maintenance and transaction exceptions. In the predictive maintenance scenario, ESM can listen for events coming from vehicles, devices, or machines. A configured Event Strategy can look for occurrences of a particular diagnostic trouble code over five days. If the Event Strategy detects the diagnostic trouble code, a Decision Strategy (using an Adaptive Model), can suggest the next best course of action to take based on the event.

Finally, a workflow will notify the customer of potential issues and provide one or several recommendations, leading to lower repair costs and more satisfied customers. For transaction exceptions, ESM can look across transactions for all customers to find any exceptions in those transactions. If the number of transactions with exceptions exceeds the threshold, ESM can start a workflow to review the exception, helping catch errors before they happen.

Pega Process AI turns data into action

Monitor and optimize every process touchpoint with our unified industry-leading process mining, decision management, and machine learning capabilities.

Learn more

Optimising the customer journey with Next-Best-Action Analytics

Infuse customer journeys with real-time decision-making.

Learn more

タグ

トピック: インテリジェントオートメーション 製品エリア: インテリジェントオートメーション

著者について

Ryan Easley is a Solutions Consultant with over 15 years of experience helping customers realize their visions with Pega.

シェアする Share via x Share via LinkedIn Copying...
シェアする Share via x Share via LinkedIn Copying...