Pick and Place: The Next Wave For Robotic Process Automation


Is it any surprise that robots have been breaking their own records as they make greater inroads into almost every industry? According to the International Federation of Robotics (IFR), the global sale of industrial robots grew by 8 percent in 2015. This translates to an addition of 240,000 robots globally. The IFR also estimates that by 2018, there will be 2.3 million operational robots in industrial use worldwide.

While robots have had a long and productive presence in industries like automotive, electrical and electronics, and metal, they have a large opportunity to flourish in industries such as banking, medical, retail and food as well. And the way I see it, the demand for robots here, will be to automate testing — a highly repetitive, lengthy process that is prone to error when done by humans. No surprisingly, Transparency Market Research (TMR), claims that the global test automation market is expected to reach $85.84 billion by the end of 2024.

But what will it look like to have robots automate the testing process. To give you an example, imagine a point of sale (PoS) or card swiping machine that banks provide to the merchants. These are small enough to fit in your hand but need to undergo over 100 tests as part of their validation before they are given out by banks. These PoS machines need to be rigorously checked for at least a couple of weeks, and should an application be updated mid-way through the testing phase, the machine needs to be validated for the entire set of applications all over again. After all, these may seem to be inconspicuous payment devices, but facilitate the transaction of hundreds of dollars each day (if not more), an error could result in an excess charge to a consumer or the transaction not happening at all, both situations could result in a negative user experience in a highly competitive market, where organizations work hard to make the right brand impact.

Robots are well suited for undertaking repetitive yet critical tasks, can be trusted to ensure higher accuracy than humans, have a lower operating cost and can be set to work with little or no training. A robot can simplify and manage the elaborate testing process accurately and cost-effectively. It can also reduce the time to market as testing can be done throughout the 24 hours cycle rather than over just an 8 hour time frame done manually by employees. Infosys is currently engaged in an innovative project on a robotic arm that will soon go live for a leading bank in the Asia Pacific region.

As I think more about robots making their way into industries like banking, medical, retail and food, other examples where robots will play a role come to mind. For example in document validation. Financial institutions mandatorily require to gather a large number of documents from consumers and much time and money is spent when personnel have the task of scanning and storing this data. Instead, a robot arm can manage this process end to end, it can physically place documents on a scanner, scan them and store them in the correct database.

Robots also have a large opportunity in the electronic industry, especially to test various functionalities of mobile phones. With a compounded annual growth rate of 8.2 percent, 1.9 billion mobile phone units are likely to be shipped out annually by 2019, and robots can play a significant role in ensuring quality control and competitive profit margins.

Another area that robots can play a critical role is in the testing of medical devices, for example pacemakers. Ensuring a pacemaker works and communicates with the physician correctly could make the difference between life and death. As the use of pacemakers and defibrillators increase, the rigor and accuracy needed to ensure the product is safe can be well introduced by robots.

Most of these examples are in the ‘pick and place’ category; where a robot can lift an object and place it in a designated place and then run a test for ‘go’ and ‘no-go’ decisions. The use of technologies such as optical character recognition (OCR) or image processing also complements the robotic arm in order to improve its accuracy.

So the next time you swipe your card at your favorite retailer; apart from thanking the customer agent, do thank that robot who tested the PoS terminal without which you could not have bought your favorite product.

For more posts, visit InfyTalk

Author: Saraswathi Thippaiah, Principal consultant, Advanced Engineering Group, Infosys Limited