Is your company struggling to forecast, and have parts arrive on time to perform preventive maintenance, or know when downtime should occur that would be least disruptive? Would your company benefit from more accurate cost predictions and a decrease in last minute emergency breakdowns? Or perhaps you cannot depend on what the system has for inventory or how many staff are required on site at a given time – there are somehow too many or not enough at any given time. These are common issues faced by many large manufacturers from Marine, to Mining to Oil & Gas, and many others. This session will provide an overview of how basic Machine Learning techniques can be applied to Supply Chain, Reliability, and Asset Management to gain increased insight, and better overall performance. But what is Machine Learning? It is a subset of Artificial Intelligence, where algorithms improve through experience (new data sets). The algorithms constantly evolve as more and more data is run through it. Machine Learning is useful for finding unknown patterns and relationships in data, such as sales, plants, store, or forecasting. It is an effective tool to gain insight and efficiency in day to day operations, while also providing a future forward view.
Danaka Porter holds a Master of Engineering in Systems and Supply Chain from MIT. She began her career at PwC in audit before moving into manufacturing. She specializes in inventory optimization and has created and copyrighted inventory algorithms for hard to predict demand situations over long periods of time. Danaka is a sessional lecturer at the University of Lethbridge’s Dhillon School of Business where she teaches Supply Chain and Project Management to upper-level students. Danaka is also a partner at a successful consulting company in Canada which takes her all over the world, and has co-founded a technology company in the US which uses machine learning algorithms she created to better match donors and recipients. Additionally, Danaka is currently doing her PhD in Cardiovascular Sciences where she is once again, using machine learning to rewrite algorithms to better highlight predictors of sudden cardiac death.