Configuring the system for running simulations of decision strategies
Version:
When you make changes to your decision data, data flows, and data models, the results of your updates impact your decision strategies. To verify what effects your changes produce on your customer-oriented campaigns, you can simulate the decision process with these updates on sample production data, including customer information, Adaptive Decision Manager input, and interaction history. After comparing the simulation results with your current outcomes, you can decide whether you want to implement your modifications.
Use case
uPlusTelco is about to release a new line of tablets. Before the launch, the marketing team is determining the demand for the new products among both their current and prospective customers. As the lead marketing strategy specialist in that team, you are responsible for designing and implementing the simulation plan for the products release. To ensure that the results are reliable, you want to use real customer data of your company to test how the offers with new products might be received, to update your global marketing strategy accordingly.
To achieve that goal, simulate how changes to decision strategies impact their outcomes by performing the following procedures:
- Configuring the data migration settings
- Optional: Configuring Deployment Manager for data migration pipelines
- Promoting an application in the production and simulation environments
- Collecting sample data for simulations
- Export and import data by doing one of the following actions:
- Automatically, by Deployment Manager: Exporting and importing simulation data automatically with Deployment Manager
- Manually: Exporting data from the production environment manually and Importing sampled data into the simulation environment manually.
- Running a decision strategy simulation on sampled production data
- Configuring the data migration settings
Configure the Dev Studio migration settings to build a framework for your simulation environment and to run tests on a sample set of migrated production data.
- Promoting an application in the production and simulation environments
Promote artifacts that you generated from the development system to the production and simulation systems. To make the process efficient, migrate your application by updating the applicable dynamic system settings (DSS).
- Collecting sample data for simulations
In configurations without Pega Customer Decision Hub, after configuring the system for data migration, complete the collection of sample data for simulations by setting up a make decision data flow. Use the following example as the baseline for your configuration. For Pega Customer Decision Hub scenarios, see .
- Exporting data from the production environment manually
In configurations without Deployment Manager, after collecting the sample data, trigger the simulation data export case type in the production environment.
- Exporting and importing simulation data automatically with Deployment Manager
You can create and run data migration pipelines in Deployment Manager to automatically export simulation data from the production environment into the simulation environment.
- Exporting data from the production environment manually
In configurations without Deployment Manager, after collecting the sample data, trigger the simulation data export case type in the production environment.
- Testing the impact of strategy design changes with simulation runs
You can use a batch case run to test the performance of your strategy as you design it, and identify which components need optimization. As you edit the strategy, run a simulation to identify the most popular propositions, check whether customers are receiving actions, and make sure that your strategy is executed as intended.
- Running a decision strategy simulation on sampled production data
After migrating the data, run simulations on the sample production data that you imported. You can use the customer sampling data set as a source to run simulations in the Pega Customer Decision Hub portal.
- Using sampled production data to test the performance of proposition filters
You can use sampled production data to improve the performance of your proposition filters by testing them against simulated audiences. In this way, you can check how many potential actions are filtered out by each component of the filter, and discover if a particular filter criterion is too broad or too narrow for your requirements.