Blogs

Optimization technology and AI in the real world: Operational efficiency stories

Content Marketing Manager: Data Science and Machine Learning, IBM Cloud and Cognitive Software, IBM

Consider this statistic from Enterprise Strategy Group (ESG):60 percent of respondents indicate that improving operational efficiency is one of the most important objectives their organization expects to achieve from their AI/ML (machine learning) investments.” This statistic was a result of looking at the top line data from ESG research that combines responses of enterprise and midmarket organizations.

Currently, many companies still have AI and ML firmly in the realm of IT, but the ESG research reveals that AI and ML are becoming more strategic to businesses. Key performance indicators (KPIs) owned by lines of business are cited as one of the top objectives for measuring the effectiveness of AI and ML. It is in this context that we would like to shine the spotlight on the role of optimization technology in supplementing ML to help businesses drive optimal business outcomes through better decisions.

In this blog post, I’ll share real-world stories of how decision optimization technology delivers prescriptive analytics capabilities and opens the door to operational efficiency. We will also introduce you to the IBM data science and AI platform solutions that can deliver operational efficiency that satisfies business needs.

How does optimization technology help decision-making?

Optimization software have at their core sophisticated optimization solvers that leverage mathematical and computational science to optimize business outcomes for use cases like predictive maintenance, price optimization, workforce scheduling, supply chain optimization, financial portfolio optimization and many more. Optimization solvers embed algorithms that can sift through all possible solutions and recommend those that are most likely to help you achieve a specific business goal—such as operational efficiency—while taking into consideration all decision variables, constraints, and tradeoffs.

As a spokesperson for a U.S. bulk carrier company explains, “Success in logistics isn’t just about having the most trucks or the best drivers. It’s also about making the smartest decisions. In many cases, you can define the optimal decision mathematically: it’s the one that maximizes certain factors, such as the utilization of trucks, trailers and drivers, while minimizing others, such as the number of empty miles driven, the amount of fuel used, and the total delivery time.”

This company used decision optimization and predictive analytics solutions with open-source Python libraries to develop a sophisticated demand forecasting model to predict incoming orders and pickup locations. The key to this kind of success is making prescriptive analytics available to data science teams.

Bringing the power of optimization to data science teams

IBM Decision Optimization for Watson Studio helps data science teams capitalize on the power of optimization software. IBM Decision Optimization is integrated with IBM Watson Studio, combining optimization and machine learning techniques with model management and deployment capabilities—and other data science capabilities. Using this solution, data science teams can drive operational efficiency and business impact across the business and optimize their ML approaches.

Data scientists can build optimization models, analyze scenarios, and solve optimization problems in the same platform, and with similar tools they currently using, to build their machine learning models.

A large bike-sharing system operator is currently using optimization software from IBM to distribute 13,000 bicycles to more than 800 stations, enabling 10,000 or more journeys per day. They calculate the optimal number of bikes for each station at any given time and plan efficient routes to help maintenance teams redistribute bikes accordingly. They solve bike inventory, distribution, and maintenance problems in seconds. And, they’ve reduced operational costs and improved performance on service level metrics.

Making optimization modeling easier while accelerating time to results

IBM Decision Optimization for Watson Studio simplifies optimization modeling by providing low-code modeling options that allow data science teams to use natural language inputs for model building. Once the models are built and validated using powerful dashboards, the model can be deployed in IBM Watson Machine Learning so business users can access the optimization models from their applications.

Take FleetPride, for example. The truck and trailer parts supplier built an optimization model of its entire distribution network. Homarjun Agrahari, Director, Advanced Analytics says that the solution helps FleetPride “work out the optimal locations where we could position a warehouse to minimize delivery time and costs across the entire network. When we’re deciding whether to build or acquire new warehouses, this kind of insight can make a significant difference to our network design strategy.”

And now companies like yours can accelerate time to results even more safely and securely thanks to IBM Watson Studio Premium Add-On for Cloud Pak for Data. IBM Cloud Pak for Data is an open, cloud-native information architecture for AI. Designed as an integrated, fully governed team platform, organizations can keep data secure at its source and add preferred data and analytics microservices as needed.

Want to learn more about applying optimization and AI for operational efficiency?

ESG recently put IBM Decision Optimization for Watson Studio and IBM Cloud Pak for Data to the test. The verdict? “If you are seeking to drive improvements in operational efficiency, optimize business decisions, transform planning processes, and perform powerful scenario modeling to evaluate and compare potential outcomes, you’d be smart to take a close look at IBM Decision Optimization for Watson Studio.”

Download the report for all the details, including how the solutions performed in specific conditions.