ASOR
•
Australian Society for Operations Research
Melbourne Chapter

 
ASOR
•
Home | Contact | Program | Newsletter | Membership | OR in Melbourne | Links | Archive

2013 Program

The default venue for the monthly lectures is
RMIT, Access Grid Room. 8.9.66

Melbourne time

Scheduled Events for 2013

DateEventSpeakerTopic
November 26 (afternoon) Workshop TBA Energy Modelling
October 16 Lecture Guy Eitzen Mathematical Applications in Transport
September 18 Lecture Marcelle Gannon Optimal Packing List for Mixed Produce Box Delivery
August 21 Student Event See program TBA
July 17 Mini-Conference See Program Recent Advances in Operations Research
June 12 Lecture Gerald Gamrath Improving Strong Branching by Domain Propagation
May 15 Lecture Olena Gavriliouk AWAT, SWAP, and AWAP: The Evolution of "Experimental'' ABS Interviewer Workforce Management Tools
April 17 Lecture AGM
April 17 Lecture David Marlow Fleet management operations research in DSTO
March 27 Lecture David Sier Modelling hospital systems: optimising patient flow and resource allocation
February 20 Lecture Andreas Ernst A tutorial on Lagrangian relaxation for integer linear programming
January 30 Lecture Ceyda Oguz Quay Crane Scheduling via Constraint Programming

Venue: RMIT Access Grid Room, 8.9.43 (Building 8, level 9, room 43)

Time: November 26 (1:30-5:30PM)

Program: Workshop

Topic: Energy Modelling

First Announcement(November 11)

The Australian Society of Operations Research (ASOR), Melbourne chapter, is pleased to announce a half-day workshop on Energy Modelling and OR, to be held on the afternoon of 26 November 2013 at RMIT and U. Newcastle, by Access Grid.

Registration is free and the theme of the workshop is energy modelling and its relationship to Operations Research as a scientific discipline.

The workshop will have three keynote presentations, one each on the topics of renewable supply modelling, energy futures modelling and demand modelling.

In addition to the keynotes we have an open call for presentations from OR students, researchers and practitioners, which can be delivered from either Melbourne (RMIT) or U. Newcastle. To reinforce the concept of a workshop, these contributed presentations will be 10 minutes long with a further 10 minutes scheduled for questions and discussion after each presentation. We strongly encourage you to present research which is incomplete/underway, related to PhD research, and/or which might even be only in the early stages of planning or commencement.

Proposals for presentations will be accepted up until Thursday 21 November. While non-presenters are permitted simply roll up on the day, we are providing some catering and so would much prefer it if you would register your interest to attend. Please send all proposals to present, and registrations of interest, to Simon Dunstall (Simon.Dunstall{at}csiro.au.)

You can just roll up on the day, but it would be better if you can indicate that you'll be attending, so we can get the a/noon tea catering right.

Program (last update: November 20)

Venue: RMIT Access Grid Room, 8.9.43 (Building 8, level 9, room 43)

Time:6:00 PM

Program: Lecture by Guy Eitzen, Aurecon

Topic: Mathematical Applications in Transport

Abstract
The field of Operations Research is designed to solve real world problems. By using a variety of different techniques and approaches to model real world behaviour, solutions can be formed and implemented. Transport is but one area where Operations Research, Mathematics and Statistics combine to undertake this goal, but is rich in diversity of problem, approach and solution. This presentation will describe a variety of real Australian projects that have involved Operations Research, Mathematics and Statistics to form tangible solutions. These will include projects ranging from crash modelling to tollway forecasting, with every freight truck, ship, quarry and bridge in between.

Bio: Dr Guy Eitzen is a transport and freight modelling specialist at Aurecon with more than 17 years of experience in consulting, particularly in mathematical modelling, optimisation, simulation and programming. He has managed large and complex freight data collection studies, has developed best practice freight vehicle estimation models throughout Australia, undertaken freight vehicle policy and intervention scenario testing, managed strategic and public transport modelling projects, and now leads the toll-road investment grade forecasting team. In 2012, he presented at the New Zealand Transport Summit, and his work in freight vehicle modelling underpins the recently released "Victorian Freight and Logistics Plan".

Prior to working for Aurecon, he obtained his passion for transport and optimisation though working with the CSIRO Mathematics and Information Sciences for Winery Optimisation (from block to bottle) for Orlando Wyndham, and Dairy Manufacturing Optimisation (from cow to cheese) for Murray Goulburn; and before that at the University of South Australia working on long haul rail scheduling problems.

Venue: RMIT Access Grid Room, 8.9.43 (Building 8, level 9, room 43)

Time:6:00 PM

Program: Lecture by Marcelle Gannon, University of Melbourne

Topic: Optimal Packing List for Mixed Produce Box Delivery

Abstract
For a university project I applied operations research techniques to a local business which home?delivers fixed?price boxes of mixed fresh produce. Currently the proprietor determines a satisfactory box contents list each week by trial and error. This project formulates a mixed-integer linear program to find an optimal list of produce types and quantities. A method is developed for approximating the customers' perception of the box quality. Implementation is carried out in Excel and, following the tuning of parameters through discussions with the proprietor, suitable box listings output. The optimisation process is fast, saving the proprietor time, but the major benefit is the ability to experiment with different price points each week, giving opportunities for increased profits and customer satisfaction.

Venue: RMIT Access Grid Room, 8.9.66 (Building 8, level 9, room 66)

Time: 4 PM, Wed August 21, 2013

Program: Student Event

On Wednesday 21 August, ASOR Melbourne will be holding our annual OR student get-together, starting at 4pm. This is an opportunity for students to socialise, meet industry practitioners and academics, and exchange views on their experiences. There will be two speakers on the day (details being finalized) as well as catered refreshments during the afternoon.

The ASOR Melbourne committee will be present, and of course the invitation to attend extends equally to our non-student members as well. For those in research and industry, this will be a way to make contact with the future of OR.

The venue is RMIT Room 8.9.66 (the Access Grid room, our usual venue for ASOR events) and it is *free*. Indeed - free ASOR membership for the rest of the year is on offer to all students who attends!

Please note that this event is not restricted to ASOR members, and that we would appreciate if members could circulate this invitation to any students that they may be supervising and working with.

For enquiries, please contact me (on Simon.Dunstall{at}csiro.au), or our ASOR Melbourne student representative Cam Macrae (on cameron.macrae{at}rmit.edu.au).

Venue: RMIT Access Grid Room, 8.9.66 (Building 8, level 9, room 66)

Time: Wed July 17, 2013

Program: One day conference

Topic: Recent Advances in OR 2013

Recent advances is a one day forum designed to give members the opportunity to exchange ideas and promote growth and activity in the OR community. Both recent interesting and/or novel applied work along with research activities are sought for this day.

Abstracts:

Abstracts of between 150 and 300 words should be submitted to the local representatives in Melbourne, Newcastle or Adelaide.

All abstracts submitted by July 15, 2013 will be published for distribution at the conference.

For Melbourne

Time and Venue
Wednesday July 17 2013
Registration 8:45 to 9:00
Conference 9am to 5pm
RMIT -- City Campus 360 Swanston St
Building 8 Level 9 Room 66
This is the Access Grid Room as in recent years the mini-conference will link with members outside Melbourne

Cost:
For those attending at RMIT to cover the cost of lunch, morning and afternoon tea plus the printing of abstracts and other administrative costs.

Members $50
Retired members$25
Student membersFREE
Non-members$55

More details, including Registration Forms:

First Call for Abstracts(PDF)

Program

8:45 - 9:10 Registration.

9:10 -10:45 From RMIT

  • Santosh Kumar: Link weight modification approaches in network routing

  • Yee Vien Ng: EngOR: Applying the principles of Hydroinformatics at City West Water

  • Niraj Ramesh Dayama*: Scheduling of heterogeneous resources for fixed tasks with sequence dependence

10:45 - 11:00 Morning Tea

11:00 - 12:30 From Newcastle

  • Jonathan Borwein: Modelling and Simulation of Seasonal Rainfall

  • Hadi Charkhgard*: A New Criterion Space Search Algorithm for Biobjective 0-1 Integer Programming

  • Mohsen Reisi*: Integer Programming Formulations for Optimizing Throughput in the Hunter Valley Coal Chain

12:30 - 1:30 Lunch

1:30 - 2:00 From Newcastle

  • Sebastian Ruther*: Integrated aircraft routing, crew pairing, and tail assignment: branch-and-price with many pricing problems

2:00 - 3:15 From Melbourne

  • Moshe Sniedovich: From lax rhetoric to irresponsible decision-making

  • Asef Nazari: Integer Programming Model for Nickel Mining with Stochasticity in Price and Exchange Rates

  • Simon Dunstall: Announcement of Student prize and close

* Student papers

Abstracts of presentations

Venue: RMIT Access Grid Room, 8.9.43 (Building 8, level 9, room 43)

Time: 4:00PM, Wed June 12, 2013

Program: Lecture by Gerald Gamrath, ZIB Berlin

Topic: Improving Strong Branching by Domain Propagation

Abstract
Deciding how to branch is one of the most important decisions any branch-and-bound based MIP solver has to take. An important component of most state-of-the-art branching rules is strong branching, which predicts the dual bounds of potential child nodes by solving auxiliary LPs.

In this talk, we describe how these predictions can be improved by including domain propagation into strong branching. We present computational experiments which show that this reduces the strong branching effort and helps to decrease both the number of branch-and-bound nodes and the overall solving time.

Venue: RMIT Access Grid Room, 8.9.43 (Building 8, level 9, room 43)

Time: 6:00PM, Wed May 15, 2013

Program: Lecture by Olena Gavriliouk.

Topic: AWAT, SWAP, and AWAP: The Evolution of "Experimental'' ABS Interviewer Workforce Management Tools

Abstract
This is a mostly non-technical talk (salted with a few equations). It conveys the story of the development and implementation of optimization, scheduling, and management tools for the Population Survey Operations (PSO) branch of the Australian Bureau of Statistics (ABS). This branch manages the interviewer workforce (500-700 people) and data capture for ABS household surveys (e.g., Labour Force Survey). The talk describes the challenges of managing a dispersed part-time workforce across several state offices with peaks and troughs in the size and location of the work program, significant workforce churn and ever-changing business rules and changes in interviewing practices. Amongst many other things, the talk advocates accessibility of scheduling, management information, business intelligence, database management, etc. tools to managerial and non-managerial levels of workforce for enduring business culture change.

Venue: RMIT Access Grid Room, 8.9.43 (Building 8, level 9, room 43)

Time: 5:30PM, Wed April 17, 2013

Program: AGM

Venue: RMIT Access Grid Room, 8.9.43 (Building 8, level 9, room 43)

Time: 6:00PM, Wed April 17, 2013

Program: Lecture by David Marlow, DSTO.

Topic:: Fleet management operations research in DSTO

Abstract
The Department of Defence within the Australian Government is being challenged to implement reform to allow it to acquire the future capabilities it needs to be a 21st Century force. The Strategic Reform Program (SRP) is aiming to reduce costs by $20 billion in a decade. Target areas for reform include Logistics and Smart Sustainment, such as focussing on improving operational availability, maintenance processes and overall efficiency and effectiveness.

The Defence Science and Technology Organisation (DSTO) has recently formed a fleet management common interest group within its Operations Research Hub. The goal of the theme is to assist Defence in prosecuting best-practice fleet management, in order to achieve fleet effectiveness targets and minimise costs for Defence and Government. This goal therefore aligns with those of the SRP.

This talk describes fleet management work and research occurring within DSTO. It provides a broad overview of areas where DSTO is involved in fleet management work across the maritime, land, air domains as well as combinations of these. This work is in areas such as fleet sizing, cost-of-ownership, sustainment, logistics, scheduling and availability modelling. Many of the problems lend themselves to OR techniques in modelling and simulation and optimisation. Complexity often arises when uncertainty is included, such as unscheduled maintenance. Specific example problems are provided, including: scheduling transport aircraft to carry loads; determining which helicopters should fly over certain periods to maximise availability; and how patrol boats can maintain continuous coverage of various areas within resource constraints.

Venue: RMIT Access Grid Room, 8.9.43 (Building 8, level 9, room 43)

Time: 6:00PM, Wed March 27, 2013

Program: Lecture by David Sier, CSIRO.

Topic:: Modelling hospital systems: optimising patient flow and resource allocation

Abstract
Hospitals now need to meet a number of patient admission targets. For example the National Emergency Access Target (NEAT) requires that by 2015 90% of emergency department patients are admitted, transferred or discharged within four hours over a calendar year, with targets around 70% to 80% in 2013/14. We describe a scenario analysis model, for simulating patients admitted to inpatient beds from the Emergency Department (ED), to look at how changing the numbers of beds in different Speciality groups at a large Melbourne hospital affected the waiting times for inpatient beds, in particular, to count the percentages of patients discharged in four hours or less from the ED to an inpatient Speciality bed. The model was controlled by an optimisation module that adjusted the number of beds in the Specialities to find the minimum number of beds needed to achieve a specified NEAT performance. We also used the model to investigate the performance of the system when common sets of beds were shared by different Specialities.

Venue: RMIT Access Grid Room, 8.9.43 (Building 8, level 9, room 43)

Time: 6:00PM, Wed February 20, 2013

Program: Lecture by Andreas Ernst, CSIRO.

Topic:: A tutorial on Lagrangian relaxation for integer linear programming

Abstract
This tutorial will cover some of the basics of Lagrangian relaxation methods used in linear programming and then move on to perhaps less well known techniques required to get good solutions from Lagrangian relaxation approaches. The focus will be on using Lagrangian relaxation in the context of (heuristically) solving integer linear programming problems. Some of the techniques discussed will be subgradient optimisation, bundle methods, the volume algorithm, Wedelin's heuristic. Finally the integration of Lagrangian methods with meta-heuristic approaches such as Particle Swarm Optimisation and Ant Colony Optimisation will be discussed. The concepts introduced in the tutorial will be illustrated using the example of multi-machine scheduling with a shared resource.

Venue: RMIT Access Grid Room, 8.9.66 (Building 8, level 9, room 66)

Time: 6:00PM, Wed January 30, 2013

Program: Lecture by Prof Ceyda Oguz, Koc University, Istanbul, Turkey.

Topic:: Quay Crane Scheduling via Constraint Programming

Abstract
This study examines the quay crane scheduling problem (QCSP) at the seaside of container terminals. A constraint programming (CP) model is constructed by taking numerous properties of the problem such as safety margins, travel times and precedence relations into account. The performance of the proposed CP model is compared with that of algorithms presented in QCSP literature. The result from the computational experiments indicates that the proposed CP model is able to produce good results while reducing the computational time. Lastly, to demonstrate the robustness and the flexibility of the proposed model, extensions of the problem with ready times and time windows are discussed.

Bio:
Ceyda Oguz is Professor of Industrial Engineering in College of Engineering at Koc University, Istanbul, Turkey. Before joining Koc University, she was a faculty member in the School of Business at The Hong Kong Polytechnic University, Hong Kong, from 1993 to 2004. She received her Ph.D. degree in Industrial Engineering from Bilkent University, Turkey, in 1993 and holds an M.S. degree (Bilkent University) and a B.S. degree (Middle East Technical University) both in Industrial Engineering. Prof. Oguz conducts research in the areas of bioinformatics, logistics, and scheduling in manufacturing and computer systems. Her expertise includes algorithm design and system modeling as well as providing optimizing and/or approximate solutions to the complex systems by means of computational methods. Prof. Oguz has participated in several research projects, which were funded by TUBITAK and jointly by the Hong Kong and the European governments, related to above fields. She has published over 30 papers in refereed journals such as Operations Research, Computers and Operations Research and European Journal of Operational Research. Prof. Oguz took part in founding the "EURO Working Group on the Operational Research in Computational Biology, Bioinformatics and Medicine" (EURO-CBBM) and acts as the secretary and the treasurer of the group since 2006.

The previous big event

IFORS 2011 Conference
July 10-15, 2011, Melbourne, Australia