DSS-2018-07: Michael Allwright and Inna Kolyshkina

Data Science Sydney proudly presents our speakers for July 2018:

Michael Allwright: "DATA FOR BAD OR GOOD? THE MINERVA COLLECTIVE"

Inna Kolyshkina: "USING DATA SCIENCE METHODS FOR PREDICTIVE ASSET MANAGEMENT FOR A LARGE AUSTRALIAN UTILITIES COMPANY. USE CASE."

200 seats available, first come - first served for members on the RSVP-yes list. Please ensure that you keep your RSVP up to date. If you cannot make it, please make you spot available for others as soon as possible.

To comply with CBA Security we need your FIRST and LAST NAME before the event. If these are not your profile name, please enter them when you register. Members who do not provide first and last name will be removed from the guest list and will not be able to attend.

Registration opens at 5:30pm and close at 6:15pm, sharp. Food and beverages between 6pm and 6:15pm and late comers cannot be admitted.

Michael Allwright: "DATA FOR BAD OR GOOD? THE MINERVA COLLECTIVE"

About the Speaker: Michael is Co-Founder and CEO of The Minerva Collective, a not for profit which uses data and data sharing for social good, and runs a data analytics consulting company focussed on using data for good. He has 12 years of experience in data science, strategy and product development, working for a mixture of startup, large and medium sized companies. Michael has a background in Pure Mathematics complemented with 10 year's consulting experience. He is also founder of Cafe Art Australia, an initiative that empowers people affected by homelessness, as well as disadvantaged indigenous through art. Michael's mission is to use data and innovation to make the world a better place.

Inna Kolyshkina: "USING DATA SCIENCE METHODS FOR PREDICTIVE ASSET MANAGEMENT FOR A LARGE AUSTRALIAN UTILITIES COMPANY. USE CASE."

About the Talk: Predictive Asset Management is becoming increasingly attractive for asset-owning organisations seek to optimise life-cycle planning, budgeting and strategic management.

This talk describes a recent project delivered to a large Australian utilities company with a geographically-diverse high-value asset portfolio. It enabled the organisation to proactively identify problematic or costly assets prior to asset failure and to use predicted future costs to inform annual maintenance planning. Intelligent allocation of limited maintenance resources and the accurate forecasting of future asset costs were the key strategic benefits of this project.

This was a challenging task because:

• The data was large, longitudinal, sparse, noisy, and sometimes incomplete and had rarely occurring outcomes (such as occurrence of an electricity outage or asset malfunction).
• The solution had to be completely transparent and auditable because the company intended to use it as the basis of funding requests from the industry regulator.
• The client also required it to be easy to understand for the senior executives and easy to implement for the IT team.

We will discuss the ways these challenges were addressed and more than twenty Machine Learning methods we compared and applied including gradient boosting, random forests, LASSO, mixed GLM and many others.

About the Speaker: Inna Kolyshkina is Director of Data Science in Analytikk Consulting Services, a Data Analytics consultancy. She has 20+ years of commercial consulting experience across a range of industries with a focus on Utilities, Insurance and Banking. Her core expertise is the use of organisational data including unstructured data (e.g. free text, images etc.) to develop evidence-based actionable insights into the business outcomes of interest.

She is the Founding Chair of the Institute of Analytics Professionals of Australia (IAPA), and regularly contributes to national and international conferences on the application of Data Science to real world problems.