During the heydays of pulp sci-fi, Robert A. Heinlein penned a now forgotten short novel titled Waldo. The parable broadly speculated on how robotics and automation would eventually come to shape the lives and the landscape of the future. Almost a century later, Heinlein’s work reads like a prophesy, foretelling the 21st century’s rapid march towards adopting machines to do men’s work. Look around and you’ll find myriad examples. Robots are putting together cars on the assembly line and acting as companions for the disabled. We are standing at the cusp of an industrial revolution that will redefine how we work, the tasks we deem worthy of human cognizance, and those we are likely to relegate as too tedious for our nimble minds.
Consider the internal processes and workflows of any sizeable corporation. Those dependent largely on human operators are likely to encounter gaps and inefficiencies, which oftentimes add up to avoidable costs and expensive inaccuracies. Some of these processes, like bills payable and collections, revolve around voluminous repetitive, rule-based, labor-intensive tasks. They not only lock in the talent pool which could be used for more intellectual endeavors, but also suffer from inherent human oversights.
Machines however excel at these very same tasks.
The Learning Curve of Robots
In this era of cognitive computing, machine learning and artificial intelligence (AI) are making headlines, mimicking the workings of a human mind to solve complex problems in every field, from medicine to manufacturing. Alongside these technologies, robotic process automation (RPA) is leaving a mark across industries.
It is unsurprisingly that enterprises generally deploy RPA to tackle high volume, rule-based, mundane, labor-intensive activities across lines of business. However, similar processes with lower degrees of standardization will stand to benefit only partially from RPA.
As for RPA’s future as an intelligent automation solution, it hinges solely on an enterprise’s overarching technology landscape. The availability of advanced computing infrastructure, robust algorithms, and advanced data storage facilities has converged to create a platform rife with potential for integrating higher forms of automation. Innovations in self-learning and self-healing solutions, natural language processing (NLP) solutions, and knowledge-driven automation can further augment business operations, while amplifying human capabilities.
In particular, Deep Neural Network (DNN) and NLP can be harnessed to infuse RPA solutions with supervised, unsupervised and, reinforcement learning capabilities. Open source libraries like Tensorflow and Theano can also accelerate implementation of techniques which include Recurrent Neural Network (RNN) and Convolutional Neural Network (CNN). No longer considered a fringe concept, JPMorgan is scanning financial contracts, a task that kept legal teams up all night, using a learning machine.
Transitioning to Intelligent Automation
. From an IT perspective, DevOps cycles are characterized by a large number of unstructured processes. Here’s where groundbreaking technology like Cognitive Insights uses machine-learning algorithms to match domain knowledge with log data, creating a data reservoir of rich insights—leading to focused solutions for issues that typically plague IT and DevOp teams.
At present, RPA and its agents operate similarly to assembly line robots – executing a defined set of sequential tasks to arrive at an outcome. In the next few years however, the picture will likely change as AI gains traction across industries. For example, RPA can be leveraged to do the ‘grunt work’ in an auto insurance claims process, retrieving DMV records and client credit reports. This information can then be analyzed by an AI platform to determine premiums, deductibles, and quotes while triggering the next steps needed to complete a sale. The next decade is expected to witness a paradigm shift with some industry leaders already toying with the idea of installing AI-powered bots on employee computers, letting them develop a program to mimic both function and process. The bot will watch the human operator over thousands of instances, learning to operate on its own – the pinnacle of cognitive automation.
RPA and AI are plugging the gaps in healthcare. Since monitoring hundreds of patients at all hours of the day is almost an impossible task, robots like the Mabu personal healthcare companion, are crucial to deal with the patient population. Robots can monitor a patient’s health and emotions, and send encrypted data to the doctor in charge.
AI and machine learning is set to find wider application in the field of medicine, especially in context of tackling even more complex challenges. For example, data scientists and students at MIT are busy leveraging natural language processing in conjunction with neural networks to capture and process millions of cancer patient records. In doing so, the research team hopes to not only build a database information retrieval tool with 98% accuracy, but also read mammograms for identifying at-risk patients.
From genetic testing to robotic surgery, or to identifying benign or malignant skin cancer, AIs and robots are making a huge impact on the healthcare industry. Questions however remain as to the ethics of machines making life and death decisions on behalf of humans.