Machine Learning: Driver of Operational Optimization

As we progress through a technological revolution and enter the age of artificial intelligence, big changes are underway for business operations. Machine learning, an application of the growing field of AI, already permeates many parts of our everyday lives, as well as increasingly informs the operations of firms around the world. Learning about AI and machine learning can feel intimidating due to the complex nature of this burgeoning field. It’s my feeling, however, that something so complicated—with layers upon layers of underlying mathematical and computational complexity, can actually be understood fairly intuitively—and utilized as another weapon in the arsenal of business optimization.

Machine learning is an application of artificial intelligence, and is made possible by big data. The collection of vast quantities of data can be exposed to algorithms, so they can come to recognize patterns, trends, and associations previously unseen. To analogize machine learning, it's essentially learning by example, rather than learning by instruction. As more and more data is collected and analyzed, the algorithms can ‘learn’ to recognize patterns with increasing accuracy. For example, the more images of a cat a program is exposed to, the more accurately and quickly it will be able to recognize an image of a cat in the future. These pattern-recognizing algorithms can be deployed to smooth business operations, and the potential for increase in efficiency is being realized in impactful ways. Take customer service bots, for example. A job traditionally done by employees of a firm, fraught with inefficiencies both for the company and customer, can now be done more cheaply, more easily, and with greater satisfaction through the use of smart bots. Large firms, in particular, have no shortage of customer interaction data regarding products sold and services provided. As aforementioned, this data can then be used to educate bots on how to provide the best service possible to customers seeking help. The more interaction they undergo, the better these algorithms become at recognizing human needs, and the speed with which they can respond and satisfy a customer is unmatchable by a human counterpart. 

This is just one example of the vast and largely untapped potential of machine learning. We are at the forefront of a big change regarding how computational advancements interact with business operations, and there are many more applications yet to be explored. As this technology becomes cheaper to implement, more broadly studied, and better understood by managers, it will shape how firms operate, and help write the future of the business world.