Decision Management (ODM,ADS)

IBM CPB4A features Business Automation Insight to collect events from business processes, decisions and other automation capabilities that derive insight into the performance of the business with dashboarding.

On the other side, Operational Decision Manager is one of the CP4BA components dedicated to business decisions automation. As part of their Business Rules management capabilities, ODM provides a Business Console where business users can author and version rule artifacts, run testing and simulation scenarios, deploy and more.

ODM simulation is a nice feature as it allows dry-run of the current version of the rules, before deploying to production. This way you can try and compare different "what-if" scenarios to anticipate the impact in the business performance before deployment. ODM shows the results of simulations as graphical KPIs in custom reports.

The challenging part of the simulation is to get good quality simulation data. This is a key point for producing significative business conclusions out of the simulation. Traditional approach is to implement product extension to connect to your corporate databases and transform historical data into simulation scenarios with the format expected as input by the decision services in ODM.

Instead of programming complex integration with corporate databases to prepare simulation data, ¿what if you can just use events stored in BAI? This is a smart approach that pays off with some benefits:

• BAI uses ElasticSearch to store events as JSON documents, and those events can be queried connectionless though HTTP REST.
• ElasticSearch is not very performant for query access, but is not required for simulation and, this way, your corporate databases keep out of this game, saving more database bandwidth for business applications.
• BAI stores the JSON input data of each execution as part of the event record. So, there is no need of data preparation, you just retrieve input data using JSON-PATH and send to the decision server to run a simulation scenario.
• There are some limitations to previous assert. Depending on your Business Rules Model, sometimes you won't be able to directly use out-of-the-box decision input data stored in BAI, and you will need some transformations. This is because of the way that BAI stores array data in the JSON document. But this is topic for a future thread. You can get further details in the IBM Community Ideas blog.
• Every BAI event stored in ElasticSearch is timestamped. Meaning that you can use timeframes in the ElasticSearch query to simulate current business rules with all the transactions registered in a similar period in previous years. For instance, if we think of a dynamic pricing decision, we can simulate with all transactions during Black Friday period in previous year.

IBM CPB4A features Business Automation Insight to collect events from business processes, decisions and other automation capabilities that derive insight into the performance of the business with dashboarding.

On the other side, Operational Decision Manager is one of the CP4BA components dedicated to business decisions automation. As part of their Business Rules management capabilities, ODM provides a Business Console where business users can author and version rule artifacts, run testing and simulation scenarios, deploy and more.

ODM simulation is a nice feature as it allows dry-run of the current version of the rules, before deploying to production. This way you can try and compare different "what-if" scenarios to anticipate the impact in the business performance before deployment. ODM shows the results of simulations as graphical KPIs in custom reports.

The challenging part of the simulation is to get good quality simulation data. This is a key point for producing significative business conclusions out of the simulation. Traditional approach is to implement product extension to connect to your corporate databases and transform historical data into simulation scenarios with the format expected as input by the decision services in ODM.

Instead of programming complex integration with corporate databases to prepare simulation data, ¿what if you can just use events stored in BAI? This is a smart approach that pays off with some benefits:

• BAI uses ElasticSearch to store events as JSON documents, and those events can be queried connectionless though HTTP REST.
• ElasticSearch is not very performant for query access, but is not required for simulation and, this way, your corporate databases keep out of this game, saving more database bandwidth for business applications.
• BAI stores the JSON input data of each execution as part of the event record. So, there is no need of data preparation, you just retrieve input data using JSON-PATH and send to the decision server to run a simulation scenario.
• There are some limitations to previous assert. Depending on your Business Rules Model, sometimes you won't be able to directly use out-of-the-box decision input data stored in BAI, and you will need some transformations. This is because of the way that BAI stores array data in the JSON document. But this is topic for a future thread. You can get further details in the IBM Community Ideas blog.
• Every BAI event stored in ElasticSearch is timestamped. Meaning that you can use timeframes in the ElasticSearch query to simulate current business rules with all the transactions registered in a similar period in previous years. For instance, if we think of a dynamic pricing decision, we can simulate with all transactions during Black Friday period in previous year.

IBM CPB4A features Business Automation Insight to collect events from business processes, decisions and other automation capabilities that derive insight into the performance of the business with dashboarding.

On the other side, Operational Decision Manager is one of the CP4BA components dedicated to business decisions automation. As part of their Business Rules management capabilities, ODM provides a Business Console where business users can author and version rule artifacts, run testing and simulation scenarios, deploy and more.

ODM simulation is a nice feature as it allows dry-run of the current version of the rules, before deploying to production. This way you can try and compare different "what-if" scenarios to anticipate the impact in the business performance before deployment. ODM shows the results of simulations as graphical KPIs in custom reports.

The challenging part of the simulation is to get good quality simulation data. This is a key point for producing significative business conclusions out of the simulation. Traditional approach is to implement product extension to connect to your corporate databases and transform historical data into simulation scenarios with the format expected as input by the decision services in ODM.

Instead of programming complex integration with corporate databases to prepare simulation data, ¿what if you can just use events stored in BAI? This is a smart approach that pays off with some benefits:

• BAI uses ElasticSearch to store events as JSON documents, and those events can be queried connectionless though HTTP REST.
• ElasticSearch is not very performant for query access, but is not required for simulation and, this way, your corporate databases keep out of this game, saving more database bandwidth for business applications.
• BAI stores the JSON input data of each execution as part of the event record. So, there is no need of data preparation, you just retrieve input data using JSON-PATH and send to the decision server to run a simulation scenario.
• There are some limitations to previous assert. Depending on your Business Rules Model, sometimes you won't be able to directly use out-of-the-box decision input data stored in BAI, and you will need some transformations. This is because of the way that BAI stores array data in the JSON document. But this is topic for a future thread. You can get further details in the IBM Community Ideas blog.
• Every BAI event stored in ElasticSearch is timestamped. Meaning that you can use timeframes in the ElasticSearch query to simulate current business rules with all the transactions registered in a similar period in previous years. For instance, if we think of a dynamic pricing decision, we can simulate with all transactions during Black Friday period in previous year.

IBM CPB4A features Business Automation Insight to collect events from business processes, decisions and other automation capabilities that derive insight into the performance of the business with dashboarding.

On the other side, Operational Decision Manager is one of the CP4BA components dedicated to business decisions automation. As part of their Business Rules management capabilities, ODM provides a Business Console where business users can author and version rule artifacts, run testing and simulation scenarios, deploy and more.

ODM simulation is a nice feature as it allows dry-run of the current version of the rules, before deploying to production. This way you can try and compare different "what-if" scenarios to anticipate the impact in the business performance before deployment. ODM shows the results of simulations as graphical KPIs in custom reports.

The challenging part of the simulation is to get good quality simulation data. This is a key point for producing significative business conclusions out of the simulation. Traditional approach is to implement product extension to connect to your corporate databases and transform historical data into simulation scenarios with the format expected as input by the decision services in ODM.

Instead of programming complex integration with corporate databases to prepare simulation data, ¿what if you can just use events stored in BAI? This is a smart approach that pays off with some benefits:

• BAI uses ElasticSearch to store events as JSON documents, and those events can be queried connectionless though HTTP REST.
• ElasticSearch is not very performant for query access, but is not required for simulation and, this way, your corporate databases keep out of this game, saving more database bandwidth for business applications.
• BAI stores the JSON input data of each execution as part of the event record. So, there is no need of data preparation, you just retrieve input data using JSON-PATH and send to the decision server to run a simulation scenario.
• There are some limitations to previous assert. Depending on your Business Rules Model, sometimes you won't be able to directly use out-of-the-box decision input data stored in BAI, and you will need some transformations. This is because of the way that BAI stores array data in the JSON document. But this is topic for a future thread. You can get further details in the IBM Community Ideas blog.
• Every BAI event stored in ElasticSearch is timestamped. Meaning that you can use timeframes in the ElasticSearch query to simulate current business rules with all the transactions registered in a similar period in previous years. For instance, if we think of a dynamic pricing decision, we can simulate with all transactions during Black Friday period in previous year.

IBM CPB4A features Business Automation Insight to collect events from business processes, decisions and other automation capabilities that derive insight into the performance of the business with dashboarding.

On the other side, Operational Decision Manager is one of the CP4BA components dedicated to business decisions automation. As part of their Business Rules management capabilities, ODM provides a Business Console where business users can author and version rule artifacts, run testing and simulation scenarios, deploy and more.

ODM simulation is a nice feature as it allows dry-run of the current version of the rules, before deploying to production. This way you can try and compare different "what-if" scenarios to anticipate the impact in the business performance before deployment. ODM shows the results of simulations as graphical KPIs in custom reports.

The challenging part of the simulation is to get good quality simulation data. This is a key point for producing significative business conclusions out of the simulation. Traditional approach is to implement product extension to connect to your corporate databases and transform historical data into simulation scenarios with the format expected as input by the decision services in ODM.

Instead of programming complex integration with corporate databases to prepare simulation data, ¿what if you can just use events stored in BAI? This is a smart approach that pays off with some benefits:

• BAI uses ElasticSearch to store events as JSON documents, and those events can be queried connectionless though HTTP REST.
• ElasticSearch is not very performant for query access, but is not required for simulation and, this way, your corporate databases keep out of this game, saving more database bandwidth for business applications.
• BAI stores the JSON input data of each execution as part of the event record. So, there is no need of data preparation, you just retrieve input data using JSON-PATH and send to the decision server to run a simulation scenario.
• There are some limitations to previous assert. Depending on your Business Rules Model, sometimes you won't be able to directly use out-of-the-box decision input data stored in BAI, and you will need some transformations. This is because of the way that BAI stores array data in the JSON document. But this is topic for a future thread. You can get further details in the IBM Community Ideas blog.
• Every BAI event stored in ElasticSearch is timestamped. Meaning that you can use timeframes in the ElasticSearch query to simulate current business rules with all the transactions registered in a similar period in previous years. For instance, if we think of a dynamic pricing decision, we can simulate with all transactions during Black Friday period in previous year.

IBM CPB4A features Business Automation Insight to collect events from business processes, decisions and other automation capabilities that derive insight into the performance of the business with dashboarding.

On the other side, Operational Decision Manager is one of the CP4BA components dedicated to business decisions automation. As part of their Business Rules management capabilities, ODM provides a Business Console where business users can author and version rule artifacts, run testing and simulation scenarios, deploy and more.

ODM simulation is a nice feature as it allows dry-run of the current version of the rules, before deploying to production. This way you can try and compare different "what-if" scenarios to anticipate the impact in the business performance before deployment. ODM shows the results of simulations as graphical KPIs in custom reports.

The challenging part of the simulation is to get good quality simulation data. This is a key point for producing significative business conclusions out of the simulation. Traditional approach is to implement product extension to connect to your corporate databases and transform historical data into simulation scenarios with the format expected as input by the decision services in ODM.

Instead of programming complex integration with corporate databases to prepare simulation data, ¿what if you can just use events stored in BAI? This is a smart approach that pays off with some benefits:

• BAI uses ElasticSearch to store events as JSON documents, and those events can be queried connectionless though HTTP REST.
• ElasticSearch is not very performant for query access, but is not required for simulation and, this way, your corporate databases keep out of this game, saving more database bandwidth for business applications.
• BAI stores the JSON input data of each execution as part of the event record. So, there is no need of data preparation, you just retrieve input data using JSON-PATH and send to the decision server to run a simulation scenario.
• There are some limitations to previous assert. Depending on your Business Rules Model, sometimes you won't be able to directly use out-of-the-box decision input data stored in BAI, and you will need some transformations. This is because of the way that BAI stores array data in the JSON document. But this is topic for a future thread. You can get further details in the IBM Community Ideas blog.
• Every BAI event stored in ElasticSearch is timestamped. Meaning that you can use timeframes in the ElasticSearch query to simulate current business rules with all the transactions registered in a similar period in previous years. For instance, if we think of a dynamic pricing decision, we can simulate with all transactions during Black Friday period in previous year.