SA Branch Meeting: Peter Kasprzak

  • 26 Sep 2019
  • 5:30 PM - 7:00 PM (UTC+09:30)
  • Engineering & Maths EM 212, The University of Adelaide, North Terrace

Branch Meeting - Thursday, 26th September 2019

The South Australian Branch of the Statistical Society would like to invite you to the September meeting of the 2019 program.

Venue: Engineering and Maths Science Building, Room EM212, North Terrace, The University of Adelaide. A campus map is available at http://www.adelaide.edu.au/campuses/northtce/.

***Please note that most entrance doors to Adelaide University buildings close at 6pm so make sure you arrive in time for the talk.

Time:

5:30pm - Refreshments in the Lecture Theatre

6:05pm - General Meeting Talk

7:30pm - A dinner will be held after the meeting at Sukhumvit Soi, 54 Pulteney St, Adelaide SA 5000. Please RSVP for dinner to aarti.gulyani@sahmri.com by 24th September as we are usually unable to change the booking numbers at the last minute.


Speaker: Peter Josef Kasprzak

Topic: Partial automation of sampling and data collection to yield better estimates for faba and canola seed emergence project.

Abstract

An often-overlooked part of experimental research is the data collection process. Errors in sampling propagate through the entire experiment, with protocols often comprising a significant portion of resources from limited budgets. The challenge is to automate sections of this process, while keeping the necessary human checks, the decrease time, manpower, and overall costs of this process. One important step is the selection of sampling protocol. Often a simple random sample can be improved upon in terms of smaller standard estimator error, and a more representative sample gathered from a smaller number of total samples. We attempt to show how simulation can be run on a representative data set to determine the optical sampling protocol for the circumstance, and at the same time re-introduce a nearly forgotten sampling technique, Ranked Set Sampling. Underutilized for 60 years since its inception at the CSIRO by G. A. McIntyre in 1952, advances in modern technology now make it feasible to add this protocol to the go-to toolbox of sampling, and realize potential gains in efficiency, with the ultimate goal of wrapping these benefits in a easy to use, field ready, minimal training required piece of software.

Biography

Peter earned a degree in Oenology from Adelaide University in 2000 and spent the next 13 years making wine in a variety of countries, for some of the largest, and smallest wineries in the world. In 2013 he returned to university and after a year of psychology he completed a B.Mathematical Sciences - hons and was offered a chance to for a Masters by research in Biometry with a focus on investigating Ranked Set Sampling in agriculture at the Biometry Hub under Dr Olena Kravchuk. Utilising UAV (drones) to aid in the data collection process with Computer Vision and Machine Learning methods he is investigating ways to make robust sampling protocols available to farmers and agronomists.


Feel free to forward this meeting notice to colleagues, all welcome.




Powered by Wild Apricot Membership Software