IBM DOC: reduce by 70% the cost of building optimization solutions
Use the new development platform from IBM to quickly create high-ROI prescriptive analytics solutions with a low TCO
By using IBM Decision Optimization Center (DOC), you can create a fully deployed optimization application in just 100 person-days thanks to:
- advanced web User Interface (UI), out-of-the-box
- integrated user management (login, access rights, workspaces)
- what-if analysis capabilities
- containerized deployment (scalability, robustness failover capabilities)
What is the real quantitative impact of DOC? How much does it accelerate the creation of applications for business users?
To measure the impact of DOC, we can take two different approaches:
- Fixed scope: fix the scope of an average optimization application and compare the development budget. Specifically, build the same solution with and without DOC, and compare how many person-days it takes to do so.
- Fixed budget: fix the development budget (e. g.100 person-days) and compare the results. More in detail, using the same number of person-days, build two solutions, with and without DOC, and compare the features and functionalities.
1. Fixed scope: DOC is 3 times faster
Without DOC, developing an average optimization solution would take approximately 300 person-days. By using DOC, the development time is reduced by 70%, to 100 person-days (see the detailed estimation in the section below).
DOC is 3 times more efficient.
We can classify development features into two main groups:
- Standard features: these are those must-have functionalities that are typically shared across all optimization solutions, such as user authentication, data input-output management, basic data visualization, what-if capabilities, DevOps activities, tests, etc.
- Customized features: these are those functionalities that are specific to each application. They typically include the optimization model, complex data visualization graphics, and specific data processing tasks.
When comparing developing with and without DOC, the time spent on customized features is the same. However, without DOC support, we would need to spend a significant amount of time to create standard features.
DOC accelerates the development of standard features by a factor of almost 12.
The table below shows the detailed time estimation on which we based our analysis.
2. Fixed budget: DOC application has a higher value and a lower total cost of ownership
With IBM DOC, we can build and deploy in just 100 person-days an average optimization application, very robust, and with all desired features and functionalities.
On the other hand, without using DOC, with 100 person-days resources, the result would be a basic and not robust application with:
- Poor data visualization (no web UI, just some Excel tables)
- No user management (login, access rights, workspaces)
- No what-if analysis capabilities, no scenario support
- Basic Docker-Compose deployment, no scalability, no failover, no robustness.
The development of custom features would be penalized as well: without a well-defined development framework, it is hard to test, identify issues, resolve bugs, and try different options. If more time is needed for testing and debugging, less time would be available for development, and the application will be poorly tested and with more bugs.
With limited functionalities and poor user experience, its adoption with business users will be more difficult, its downtime will be high and its overall maintenance over time will be more costly. In a few words, the total cost of ownership for the company will be significantly higher.
With the same overall budget, the application built with DOC would deliver a significantly higher value to the business and would represent a much lower total cost of ownership.
In 100 person-days of development without DOC, we obtain a low-quality application that will end up having a higher total cost of ownership (TOC).
The table below shows the detailed time estimation on which we based our analysis.
Conclusions
In this article, we have shown that:
Decision Optimization Center accelerates the development of full solutions: it accelerates the development of standard features by about 90%.
This allows you to focus on the specific components of your application — those that require more specialized skills, such as the optimization model and the data model design.
Thus, at equal development and deployment cost, DOC applications deliver higher business value.
Plus, they have a lower TCO: thanks to the optimization server, they are easy to extend and allow for CPUs and Licences sharing; thanks to containerized technology, they are stable and cheaper to maintain, and thanks to the use of open-source libraries, their UI is easy to extend and adapt.
Finally, a couple of “soft” considerations:
- If all of your applications share the same structure, the development and deployment practices can be standardized, increasing efficiency.
Working on multiple projects at the same time will be less disruptive, and fixing a bug in an old application will not be a nightmare anymore. - Delivering — at the same budget — a more polished and professional application increases end-users’ acceptance and adoption. No matter how advanced a solution is, if it's not adopted by end-users, it is worthless.
- You often spend a lot of time developing features that ultimately turn out to be useless. DOC allows for frequent interactions with business users, making them part of development decisions. With DOC, in a matter of hours, you can build a Minimum Viable Product around your optimization model, put it in the hands of your business users, and progressively improve it based on their feedback.
Are you ready for your next DOC application? Get started!
IBM® Decision Optimization Center provides a configurable platform for building, deploying, and monitoring decision support applications for business users.
Stay up-to-date with our announcements and tutorials!
Learn more about IBM Decision Optimization Center
Download DOC datasheet
Book a demo with an expert
