7 months ago

TRAMSCOOP_final cut

Industry Connect

Industry Connect Catching up PAGE 10 with Automation Automation – referred to by some as the fourth scientific revolution after the Industrial, Electricity and the Internet, is the process of getting robots and computers to do the repetitive and difficult stuff. Let’s have a look at some of the key aspects followed by an interview with Adrienne Hill – VP, RPA – Automation CoE Current Trends : Robotic Process Automation (RPA) – essentially just software that is configured to perform tasks human’s normally perform using a set of standard rules. Execution is abstracted from view. Robotic Desktop Automation (RDA) – Same as RPA but executed on the CCP or end user’s computer. Typically used to help pull or push data to the multiple applications the user is working with to completed a single process. RPA and Integration with Other Tools/Services – Use of RPA or RDA alongside other tools that help enhance the capability. An example is OCR – the robot can extract text from images using an OCR – enables it to be able to automate processes that have a requirement to read scanned semi-unstructured or structured documents. Another example is using RPA and interfacing with tools that provide intelligent capabilities such as artificial intelligence (AI) or machine learning (ML). Intelligent/Cognitive Automation – essentially applies to all of the above but with some elements of machine learning applied to it – which in simplistic terms means the automation will be able to, based on large amounts of data, start to make it owns decisions using probability.

Industry Connect PAGE 11 On the horizon – Artificial Intelligence/Machine Learning and More: Autonomics – Self learning engines. Autonomic computing is not a new term, it came from IBM in the early 2000s. It refers to self-regulating or self-healing systems. RPA could use autonomics to better handle anomalies in a process or recover from an exception based on a set rules that govern it. Machine Learning (ML) and Deep Learning – Algorithms can be trained to search through data , group and cluster data and arrive at certain decisions. Virtual Agents – Having meaningful and effective interactions with customers – relevant conversations with customers. Examples: IPSoft, Amazon Alexa and Google Voice Assistant. Computer Vision – Exists in self driving cars and our phone camera’s today – think facial detection when taking a picture. As this technology gets better there will be many applications for robots to better interact with the world around them. Neural Networks - Algorithms to detect patterns. Applicable mostly in the area of robotics to how it can be used to cluster and organize unlabeled and unstructured data. Also used alongside deep learning to establish correlations - this can be applied to predictive analytics. Natural Language Processing (NLP) – A computer’s interpretation of human language in a useful way. Intelligent chat bots might use NLP to determine sentiment or the tone of the language in a text message or spoken word.