What's new and changed in Decision Optimization

Decision Optimization updates can include new features, bug fixes, and security updates. Updates are listed in reverse chronological order so that the latest release is at the beginning of the topic.

You can see a list of the new features for the platform and all of the services at What's new in Cloud Pak for Data.

Installing or upgrading Decision Optimization

Ready to install or upgrade Decision Optimization?

Related documentation:

Refresh 9 of Cloud Pak for Data Version 4.0

A new version of Decision Optimization was released in May 2022.

Operand version: 4.0.9

This release includes the following changes:

New features

Deprecation notice for do_12.9 and do_12.10 CPLEX runtimes

The CPLEX runtimes do_12.9 and do_12.10 will be removed in the next release of Cloud Pak for Data, so you must start using the do_20.1 runtime.

Important: Make sure that all your existing deployments use do_20.1 before you upgrade to the next release of Cloud Pak for Data. Any outdated deployments will be removed during the upgrade to the next release.

You can update your existing deployments by using either of the following options:

REST API
UI

Security fixes

This release includes fixes for the following security issues:

CVE-2022-27191, CVE-2022-24785, CVE-2022-1271, CVE-2022-1154,

CVE-2021-45346, CVE-2021-3807, CVE-2021-23841

See https://exchange.xforce.ibmcloud.com/ for details.

Refresh 8 of Cloud Pak for Data Version 4.0

A new version of Decision Optimization was released in April 2022.

Operand version: 4.0.8

This release includes the following changes:

New features

The 4.0.8 release of Decision Optimization includes the following features and updates:

The default Python for Decision Optimization users is now 3.9. Python 3.7 and 3.8 have been removed.

Python is used to run Decision Optimization models formulated in DOcplex (a native Python API for Decision Optimization) in both Decision Optimization experiments and Jupyter notebooks. Modeling Assistant models also use Python because DOcplex code is generated when models are deployed.

Models that are formulated in OPL (a modeling language), or in specific file formats for CPLEX or CP Optimizer (the solver engines), such as LP or CPO formats, are not affected by this Python update.

If your models use the default Python version no change is required because the upgrade will be handled automatically. If however you have explicitly specified Python 3.7 or 3.8 in your model, you must update this version specification.

For experiments, you can select the Python version in the Run configuration pane. (For details, see Run parameters.) If the Python version parameter is not specified in your experiment, the default version is used automatically, and no action is required.
For Jupyter notebooks that use older versions of Python, you can change the Python version by selecting Change environment in the list of actions. Before you can edit your notebook, you must select a supported environment. When your notebook is open for editing, you can select an environment definition on the Environment tab in the Information pane in your notebook. For details, see Decision Optimization notebooks.
If you have an existing deployed model that explicitly uses Python 3.7 or 3.8 you must take one of the following actions:
- Edit and redeploy your model. For details, see REST API example.
- Change or remove the Python version by using the REST API without having to redeploy the model. For details, see Changing the Python version for an existing deployed model using the REST API.

Bug fixes

This release includes the following fixes:

Issue: The full CPLEX commercial edition was not working in IBM runtime 22.1 on Python 3.9.
Resolution: This is now resolved.

Security fixes

This release includes fixes for the following security issues:

CVE-2022-25315, CVE-2022-25314, CVE-2022-25313, CVE-2022-25236, CVE-2022-25235, CVE-2022-23852, CVE-2022-23308, CVE-2022-23219, CVE-2022-23218, CVE-2022-22827, CVE-2022-22826, CVE-2022-22825, CVE-2022-22824, CVE-2022-22823, CVE-2022-22822, CVE-2022-0413,, CVE-2022-0392, CVE-2022-0361, CVE-2022-0359, CVE-2022-0318, CVE-2022-0261,

CVE-2021-46143, CVE-2021-45960, CVE-2021-44577, CVE-2021-44576, CVE-2021-44575, CVE-2021-44574, CVE-2021-44573, CVE-2021-44571, CVE-2021-44570, CVE-2021-44569, CVE-2021-44568, CVE-2021-43618, CVE-2021-4209, CVE-2021-40528, CVE-2021-3999, CVE-2021-3807, CVE-2021-3634, CVE-2021-35939, CVE-2021-35938, CVE-2021-35937, CVE-2021-31566, CVE-2021-29469, CVE-2021-23177,

CVE-2020-21674, CVE-2019-12904,

CVE-2018-16428, CVE-2018-1000880, CVE-2018-1000879, CVE-2018-1000654,

CVE-2017-14501, CVE-2017-14166

See https://exchange.xforce.ibmcloud.com/ for details.

Refresh 7 of Cloud Pak for Data Version 4.0

A new version of Decision Optimization was released in March 2022.

Operand version: 4.0.7

This release includes the following changes:

New features

The 4.0.7 release of Decision Optimization includes the following features and updates:

Deprecation notice for Python 3.7 and 3.8: Python 3.7 and 3.8 is deprecated and will be removed in an upcoming refresh. Start using Python 3.9.

Security fixes

This release includes fixes for the following security issues:

CVE-2022-21668, CVE-2022-0530, CVE-2022-0529,

CVE-2021-43803, CVE-2021-43519, CVE-2021-4217, CVE-2021-41092, CVE-2021-3997, CVE-2021-3981, CVE-2021-39293, CVE-2021-39031, CVE-2021-38297, CVE-2021-38185, CVE-2021-36221, CVE-2021-3408, CVE-2021-29923,

CVE-2020-35512,

CVE-2019-8906, CVE-2019-8905, CVE-2019-7317, CVE-2019-16866

CVE-2018-20839, CVE-2018-1121

CVE-2009-0733, CVE-2009-0723, CVE-2009-0581

See https://exchange.xforce.ibmcloud.com/ for details.

Refresh 6 of Cloud Pak for Data Version 4.0

A new version of Decision Optimization was released in February 2022.

Operand version: 4.0.6

This release includes the following changes:

New features

The 4.0.6 release of Decision Optimization includes the following features and updates:

New Python version 3.9: Python 3.9 is now supported for building and deploying Decision Optimization models. The default version remains 3.8. For details, see Decision Optimization notebooks. To use Python 3.9 in your Decision Optimization experiment, select this version in the Run configuration parameters drop-down menu. If you do not select a Python version, the default version 3.8 is used. For details, see Run configuration.
To modify or redeploy any existing deployed Python or Modeling Assistant models to use Python 3.9, use the oaas.docplex.python parameter to specify Python version 3.9.
New Python client library: When you upgrade from Cloud Pak for Data 3.5, the old Python client library dd-scenario is replaced by the decision_optimization_client Python library. You must update any 3.5 notebooks that you still plan to use, if they use the old Python client library.

Security fixes

This release includes fixes for the following security issues:

CVE-2021-44832, CVE-2021-44228, CVE-2021-43797, CVE-2021-42574, CVE-2021-4019, CVE-2021-3984, CVE-2021-3918, CVE-2021-3872, CVE-2021-3801, CVE-2021-3796, CVE-2021-3778, CVE-2021-37750, CVE-2021-37713, CVE-2021-37712, CVE-2021-37701, CVE-2021-3749, CVE-2021-37137, CVE-2021-37136, CVE-2021-3664, CVE-2021-36222, CVE-2021-36090, CVE-2021-36087, CVE-2021-36086, CVE-2021-36085, CVE-2021-36084, CVE-2021-35942, CVE-2021-35517, CVE-2021-35516, CVE-2021-35515, CVE-2021-3541, CVE-2021-3537, CVE-2021-3521, CVE-2021-3520, CVE-2021-3518, CVE-2021-3517, CVE-2021-3516, CVE-2021-3445, CVE-2021-3426, CVE-2021-33574, CVE-2021-33560, CVE-2021-32804, CVE-2021-32803, CVE-2021-32796, CVE-2021-32723, CVE-2021-3200, CVE-2021-29425, CVE-2021-28153, CVE-2021-27645, CVE-2021-27218, CVE-2021-23840, CVE-2021-23490, CVE-2021-23445, CVE-2021-23343, CVE-2021-22925, CVE-2021-22924, CVE-2021-22898, CVE-2021-22876, CVE-2021-20266, CVE-2021-20232, CVE-2021-20231,

CVE-2020-8908, CVE-2020-36048, CVE-2020-24370, CVE-2020-16135, CVE-2020-14155, CVE-2020-13435, CVE-2020-12762,

CVE-2019-5827, CVE-2019-20838, CVE-2019-19603, CVE-2019-18218, CVE-2019-17595, CVE-2019-17594, CVE-2019-13751, CVE-2019-13750

See https://exchange.xforce.ibmcloud.com/ for details.

Refresh 5 of Cloud Pak for Data Version 4.0

A new version of Decision Optimization was released in January 2022.

Operand version: 4.0.5

This release includes the following changes:

Security fixes

This release includes fixes for the following security issues:

CVE-2021-45105, CVE-2021-45046, CVE-2021-3421, CVE-2021-33910, CVE-2021-22947, CVE-2021-22946,

CVE-2020-27619, CVE-2020-1712, CVE-2019-20386,

See https://exchange.xforce.ibmcloud.com/ for details.

Refresh 4 of Cloud Pak for Data Version 4.0

A new version of Decision Optimization was released in December 2021.

Operand version: 4.0.4

This release includes the following changes:

New features

The 4.0.4 release of Decision Optimization includes the following features and updates:

Support for Power® 9 hardware: You can now install Decision Optimization on Red Hat® OpenShift® Container Platform Version 4.8 clusters running on Power 9 hardware.

Security fixes

This release includes fixes for the following security issues:

CVE-2021-3572, CVE-2021-33938, CVE-2021-33930, CVE-2021-33929, CVE-2021-33928, CVE-2021-33503, CVE-2021-3114, CVE-2021-29921, CVE-2021-27219, CVE-2021-23336, CVE-2021-22923, CVE-2021-22922, CVE-2021-20271,

CVE-2018-16429,

CVE-2017-18018

See https://exchange.xforce.ibmcloud.com/ for details.

Refresh 3 of Cloud Pak for Data Version 4.0

A new version of Decision Optimization was released in November 2021.

Operand version: 4.0.3

This release includes the following changes:

New features

The 4.0.3 release of Decision Optimization includes the following features and updates:

CPLEX® and CPO models support in Decision Optimization experiments (Watson Studio)

You can now import and solve CPLEX and CPO files in Decision Optimization Experiments. For details, see Build model view.

Decision Optimization experiment UI improvements (Watson Studio)

You can more easily select how to formulate your model in the Decision Optimization Experiment UI. For details, see Build model.
You can now zoom in on the solution graph in the Explore Solution view.

Deprecation notice for CPLEX 12.9

The CPLEX 12.9 model type is deprecated in Watson Studio andWatson Machine Learning and will be removed in an upcoming release. Migrate to the latest version, CPLEX 20.1.

For details on Decision Optimization model types, see Model deployment.

Security fixes

This release includes fixes for the following security issues:

CVE-2008-5358, CVE-2008-5352, CVE-2008-5349, CVE-2008-5347, CVE-2008-3110, CVE-2008-3109, CVE-2008-3106, CVE-2008-3105, CVE-2008-3103, CVE-2008-1191,

CVE-2007-3716

See https://exchange.xforce.ibmcloud.com/ for details.

Refresh 2 of Cloud Pak for Data Version 4.0

A new version of Decision Optimization was released in October 2021.

Operand version: 4.0.2

This release includes the following changes:

New features

The 4.0.2 release of Decision Optimization includes the following features and updates:

Support for Git-based projects: You can now use Decision Optimization experiments in Watson Studio Git-based projects.
Python 3.8: Python 3.8 is now the default version for Decision Optimization experiments. Python 3.7 is deprecated but still supported.

Security fixes

This release includes fixes for the following security issues:

CVE-2021-35065, CVE-2021-22918, CVE-2020-25658

See https://exchange.xforce.ibmcloud.com/ for details.

Refresh 1 of Cloud Pak for Data Version 4.0

A new version of Decision Optimization was released in August 2021.

Operand version: 4.0.1

This release includes the following changes:

New features

The 4.0.1 release of Decision Optimization includes the following features and updates:

Support for upgrade

You can now upgrade Decision Optimization from the following Cloud Pak for Data releases:

Cloud Pak for Data Version 3.5.x
Cloud Pak for Data Version 4.0.x

Sort data tables

You can now sort data tables in the Prepare Data view.

Support for Python 3.8

By default, Decision Optimization experiments use Python 3.7. However, you can edit the run parameters for your experiment to use Python 3.8 instead.

Bug fixes

This release includes the following fixes:

Issue: Importing and exporting Decision Optimization assets were not supported in Cloud Pak for Data 4.0.
Resolution: This is now resolved.

Security fixes

This release includes fixes for the following security issues:

CVE-2019-4441, CVE-2019-4305, CVE-2019-4304, CVE-2019-4046,

CVE-2018-1902, CVE-2018-1851, CVE-2018-1683, CVE-2018-1553,

CVE-2016-0378

See https://exchange.xforce.ibmcloud.com/ for details.

Initial release of Cloud Pak for Data Version 4.0

A new version of Decision Optimization was released as part of Cloud Pak for Data Version 4.0.0

Operand version: 4.0.0

This release includes the following changes:

New features

Version 4.0.0 of the Decision Optimization service includes the following features and updates:

New Decision Optimization runtime

When you run a model in a Decision Optimization experiment, the new do_20.1 runtime is used by default. For details, see Run model view.

Support for C# models

You can delegate the Decision Optimization solve to run on Watson Machine Learning .NET (CPLEX or CPO) models. For details, see Delegating the Decision Optimization solve.

New features in the Modeling Assistant

The Modeling Assistant now provides support for:

User-defined decisions

You are no longer restricted to only using decisions deduced from your intent. You can now define your own decisions using the advanced settings and decision tabs, where you can select your decision type and its dimensions (data table or column). You can then configure new rules and objectives which use your newly defined decision.

For details, see Defining custom decisions

Multi-concept iteration

You can specify new groups of rules in natural language by combining different concepts and iterating over these combinations. For example, you can combine employees and days and then state that you want your rule to apply to each employee-day combination.

For details, see Using multi-concept iteration

Logical constraints and associated concepts

You can specify that if one constraint applies, then another constraint also applies. You can also express certain conditions in constraints more concisely and intuitively using the word associated in your natural language expression. This automatically makes the necessary logical connection between the concepts you are referring to, without you having to use more complicated join expressions.

For example, this constraint illustrates both logical constraints (if.. then) and the associated keyword in the Modeling Assistant natural language:

For each employee-day combination, 
  if (the number of assignments of Employee is equal to 0) 
    then (the number of associated Oncall duties is equal to 0 )

For details, see Using logical constraints.

For a video demonstrating these new Modeling Assistant features, see Use Decision Optimization Modeling Assistant video.

CPLEX V.20.1

CPLEX V.20.1 is now available in Watson Machine Learning. For details, see Model deployment.

Support for audit logging

Decision Optimization integrates with the Cloud Pak for Data audit logging feature. Events related to Decision Optimization experiments, scenarios and solves now generate audit records. For details, see Services that support audit logging.