This post is a guest post from Dr Andreas Kempa-Liehr, a data scientist who is one of the newest members of academic staff in the Department of Engineering Science.

Decisions under uncertainty

The only certain thing about the future is its uncertainty. Yet we are making decisions for the very next future, both in our private life and the business/engineering processes, we are responsible for. The enablers for these decisions are our very individual skills, which we have learned from interactions with our environment. This kind of knowledge can be interpreted as our very personal, intrinsic model of the environment, which we are using for solving problems. It comprises both our expectation of what is likely to happen and the understanding of how to achieve the desired outcome.

The problem is that people are not very good in making decisions under uncertainty, which might be boiled down to the following quote of Amos Tversky, who worked with Nobel-prize winner Daniel Kahneman [1] on the discovery of systematic cognitive biases:

“The evidence reported here and elsewhere indicates that both qualitative and quantitative assessments of uncertainty are not carried out in a logically coherent fashion, and one might be tempted to conclude that they should not be carried out at all.” [2]

Does this mean, that objective algorithms should be able to make better microdecisions? Yes, but for implementing them one needs a clear understanding on what the meaning of better is (Domain Expertise) in order to develop models for predicting the information needed for doing better (Data Science) and models for making decisions from the provided information (Operations Research). The critical part is the mathematical interface between predictive model and decision model, which should not be a single number of a predicted outcome (point estimate) but a probability for each possible outcome given the actual circumstances (conditional probability distribution). The important point is that conditional probability distributions allow to systematically take into account the uncertainty of the predictions such that cost-optimal decisions under uncertainty can be made.

Automating Micro-Decisions

Have a look at the following slide, which has been captured from a presentation of M. Michaelis given at the 4th Big Data & Analytics Congress [3]. It shows the out-of-stock rate of 10 stores, which had their replenishment processes being switched to a data driven approach based on conditional probability distributions for expected sales. In the beginning the suggested replenishment orders could be altered by staff, but after a transition period the processes were switched to full automation. The slide is in German, but the diagram speaks for itself: It shows the plummeting of the out-of-stock rates after switching to fully automated replenishment orders.




[1] D. Kahneman. Thinking, Fast and Slow. Farrar, Straus and Giroux, New York, 2011.

[2] Amos Tversky and Derek J. Koehler. Support theory: A nonextensional representation of subjective probability. Psychological Review, 101(4):547–567, 1994.

[3] Mark Michaelis. Case Study Kaiser’s Tengelmann: Prognoseverfahren im Dispositionsumfeld.






One of my favourite things about the month of March is welcoming the new group of Engineering Science and Biomedical Engineering undergraduate students into the Department.  Their energy and enthusiasm is always refreshing.  As usual at the end of the second week of semester everyone piled onto three buses for a field trip. The trip started with visits to companies who employ our graduates.  We often catch up with former students (now launching successful careers) on those visits.  We all ended up Rotorua (after a great visit to Contact Energy's Wairakei geothermal plant in Taupo).

This year I added a new component to the trip with a visit toTe Puia.  The group were welcomed onto the Te Puia marae, enjoyed a cultural performance and a hangi dinner, and then witnessed the awesome sight of the Pohutu geyser discharging as the sun set.

Once we returned to the backpacker accommodation that was home for the night I talked to the students about why the Te Puia visit was part of the trip.  As the students start their journeys through our degrees I want them to remember that NZ operates under a principle of partnership - through the Treaty of Waitangi.  We talked about the fact the Contact Energy and the Tauhara North No.2 Trust work as partners on the geothermal developments the students had seen that morning.

Kia mau Ki te whenua (hold fast to the land).
Whakamahia te whenua (make use of the land).
Hei painga mo nga uri whakatipuranga (for the future generations).

We talked about the key role of engineers in supporting sustainability - an example of which is the computer modelling work done in the Department of Engineering Science that considers the geothermal resource underlying the city of Rotorua.  That model helps understand the impact of the closure of private bores in Rotorua which allowed the important Pohutu geyser begin to flow again.

For the next few years our new students will face the challenge of building their knowledge of a set of mathematical and computational tools that can help understand natural systems (as well as manufactured ones).  However one challenge I would like to see engineers and geoscientists embrace is broadening their insights to acknowledge and value Matauranga Maori.  Dan Hikuroa recently discussed this in a geological context on Maori TV - pointing out that events attributed historically by Maori to taniwha may well be attributable to earthquakes.   There's definitely insight to be gained if everyone - regardless of their cultural heritage - integrates all forms of knowledge of the processes and forces that shape the earth.  Those processes give us geothermal energy reservoirs which sustain us, and earthquakes which we must be resilient to.




The Geothermal Institute are currently hosting a group of Indonesian and Phillipino geothermal energy professionals (from a wide range of disciplines) for a 4 week project management course.  This course is being as a Ministry of Foreign Affairs and Trade sponsored initiative.  On Day 1 I wanted to "break the ice" (though the group are warm and friendly) and get the course participants working in teams.  The course has plenty of time for "techy" group work so instead of a task with a geothermal focus I set everyone the "Marshmallow" challenge.  This requires a group to build a structure to support some marshmallows using (dry!) spaghetti, and adhesive tape.  Some versions also offer the participants some string - but none was to be found in my kitchen cupboards the night before!  An outline of the challenge set up can be found here.

Some of the structures that resulted look like this.

Clearly specifying requirements matters in any project.  I had forgotten to mention that the structures needed to be freestanding - so this group cleverly took their structure to the ceiling.

Group dynamics in the marshmallow challenge is the subject of a TED talk by Tom Wujec.  So how do teams of various kinds do?  Unsurprisingly it depends on the skills, and the mix of skills in the team.  Tom Wujec's talk compares the performance of teams with different backgrounds in this graphic.

When I revealed this image to our course participants they found the first few bars entertaining!   Personally I enjoyed seeing the finding that teams which are a mix of CEOs and Executive Admins outperforms teams which are only have CEOs.  The organisation and facilitation skills Executive Admins bring are a very important part of delivering on the project goal.  I know the work I do really benefits from the professional staff around me who diversify the skill mix in the Department and the Institute.

But why did the young children do so well?  They experiment and prototype naturally - allowing them to test assumptions.  That supports innovation and creativity.  For more thoughts on being curious culturing creativity there's further discussion and advice here.











I haven’t been blogging for a while – and I have missed taking the time to jot down some thoughts on things that are happening in my various roles at the University.  The end of the summer has been a busy time – lots of grant writing, and preparations to bring in a new class of both undergraduate and graduate students.

This week I’d like to focus on our new group of postgraduate students in the Master of Energy program.  The program is an interdisciplinary program for students from Science, Engineering and Business.  The program had around 40 students in various phases of their degree right now.

This week I’ve really enjoyed meeting the new intake of students.  They are all extremely energetic (excuse the pun!) and have a real hunger for knowledge that they truly hope to “change the world” with.  The majority of the students came from overseas – and literally come from every corner of the globe.  We have students from North America, South America, Asia, Africa, Europe and the Pacific all in class together.

I travel a lot and believe that global problems – such as improving access to clean energy worldwide – are best solved through global collaborations.  We kicked those collaborations off at an icebreaker event last week where the students worked on building models of energy related devices (such as wind turbines) from some kit-sets.  I look forward to seeing what ideas the students build while they are with us.

I enjoy the fact that the staff and students I work with have tremendously diverse backgrounds but share a passion for common scientific questions.  I am also proud that NZ’s Ministry of Foreign Affairs and Trade lists renewable energy as a priority area for their scholarship funding.  That means some of the students in our new cohort are supported through this scholarship scheme – which is then a mechanism for exporting Kiwi energy “know how” offshore.

Access to reliable/affordable electricity is transformative in society.  Most (but not all) New Zealanders can take that their access to grid-connected electricity for granted.  Locally I was part of a team late last year which reflected on the state of the nation (in particular in my section access to clean energy) as part of the Habitat III report to the United Nations.  Sharing ideas and experience via organisations such as the UN is important.  However our postgraduate program is like a mini UN on a daily basis!




Prof. Marcus du Sautoy said "Mathematics can often appear arcane, esoteric, unworldly and irrelevant."   In that New Statesman article Prof. du Sautoy then went on to counter that and outline his views on relevance and importance of mathematics.  (Aside - [If you're into mathematics/physics/Hitchhikers Guide to the Galaxy - you may also enjoy Prof. du Sautoy's thoughts on why 42 is in the fact the "Answer to the Ultimate Question of Life, the Universe, and Everything" published here.)

Like Prof. du Sautoy I believe mathematics offers us tremendous insight into the way the world works.  In the work we do in the Department of Engineering Science we describe the process of using mathematics to understand the world as "mathematical modelling".   So what is mathematical modelling? Wikipedia gives a definition of  "A mathematical model is a description of a system using mathematical concepts and language."   While I agree that's a valid definition - if we met at a cocktail party and I told you that was what I do, you may be politely looking for ways to break off the conversation (depending on your level of interest in mathematics).

The word "language" in the Wikipedia definition is however very important to me.  When I was at school I loved learning languages (and still do).  I studied French to 7th form (now called Year 13) and Latin to Year 12.  I frequently comment that for me mathematics is in many ways just another language.   The mathematical modelling process involves a translation of a problem that arises from a community, industry, science, government etc. into a set of mathematical statements that capture the relevant details.

Once translated into mathematical language do I end up with a set of mathematical equations that I can solve by hand (on paper)?  Not usually!  I sometimes say in jest that while I am a reasonable mathematician, I specalise in writing down equations I can't solve.   This means the modelling process typically includes a phase of translating the mathematics involved into a form that can be solved by a computer.  All going to plan the computer-based version of the model then becomes a "crystal ball" where the modeller can ask "what if?" questions to explore uncertainty (e.g. in a traffic flow model what happens if 50% more cars per hour travel on a certain road due to a special event in the area?)  To make meaningful predictions of the future behaviour of a system the model must be validated against previous observations of that system (e.g. if I want to predict future flow rates and temperatures in a geothermal well, my model should ideally be able to retroactively recover previous flow behaviour).

Finally modellers need to be able to address the "Why?" question ... why was the model constructed?  Does it provide a robust answer to that question?  Through the modelling process having an appreciation of the topic being addressed is very helpful.  Within the Engineering Science degree we allow students to build their understanding of various domains where modelling may be applied (from financial markets to environmental engineering) by having the flexibility to include electives from outside the Department throughout the degree.  We hope that makes them better mathematical modellers!







Like many New Zealanders I was (till July last year) the owner of an older car (I won't say how old!) that had a lot of kilometres on the clock.  When it came time to change vehicles I decided to "walk the talk" and buy an electric car.  I took the plunge and bought a second hand Nissan Leaf - and it was "love at first drive".  The car is very quiet, accelerates beautifully and is generally fun to drive. There's no petrol engine in the car, and it charges overnight (in about 6 hours) from a household plug (with a higher current "caravan" socket).  So does "driving electric" make sense?  There were about 1,250 electric vehicles registered in NZ when I bought mine - today there are more like 2,250 (see www.driveelectric.org.nz for current stats).  That means more and more NZers think it does make sense.

Nissan Leaf vehicles charging at a Vector charging station in Takanini. Mine was headed on a day trip further south that day.

I was pleased to see research published by EECA in New Zealand that confirmed the environmental benefits of electric cars.  They state:

"Across the lifecycle, pure EVs have around 60% fewer CO2 emissions than petrol vehicles. When we just look at the CO2 emissions from use, New Zealand’s high proportion of renewable electricity generation means EVs have around 80% fewer CO2 emissions when driven in New Zealand.  As the renewable proportion of New Zealand’s electricity continues to grow, the CO2 emissions from an EV will reduce further."

EECA's research also dispels concerns re net environmental impacts associated with lithium production for electric vehicle batteries.

So what's the catch?  The typically advertised range of my model of Nissan Leaf, with its 24 kWh battery capacity, is 125 km on a full charge.  That assumes driving on the flat on smooth roads etc. - so in reality I get less commuting range than that since we live at the top of a long winding hill.  I have a fairly significant commute so I charge the car every night at home.  What happened to my power bill?  I opened the first one with bated breath!  However the car's energy demands are relatively modest - so I traded filling up on a full tank of petrol every week, for an increase of $20 to $30 on my monthly power bill (I have a discounted rated for night rate electricity, and also have solar panels on the roof of my home that help charge the car when it's home at the weekends).

Colleagues in the Department are interested in supporting NZ's transition to more electric vehicles.  We've had a student project building web-based mapping tools that assess the viability of the use of an electric vehicle in Auckland for commuting taking into account speed limits and terrain (note that driving downhill regenerates charge in the car battery).  The tool is not quite ready for public use but I looked forward to seeing it deployed.

Companies such as Vector are deploying fast chargers that will charge a car like mine in 20 minutes or so.  Their chargers are currently free for public use, but that will change at some point.  The Government, via EECA are also trying to accelerate the transition to  electric vehicles, via the Low Emission Vehicles Contestable Fund.

So, a longer post than usual, but I have a lot of love for driving electric!






Happy New Year!  Are you happy to be back at work, or looking forward to being back in class in 2017?  Do you know why?

In 2016 I served on various panels and committees (on campus and beyond) in which decisions were made that offered people or organisations opportunities and resources, or that looked to make decisions on process/structure in organisations.  I won't elaborate on the details since it many cases they are confidential.  However when I reflect on the most successful interactions they were normally with people who had clearly answered the "why" question for themselves and let that show.


The "why" that gets me out of bed in the morning is the the belief that engineering mathematics is a toolbox that can offer businesses, government and communities insight and novel solutions to the problems they face - in turn creating value for shareholders and society.  As a Department Head I aim to create an environment where students can develop their expertise with those mathematical tools (and create new tools). Anyone shadowing me in my office would see "what" I do in business hours - meetings, financial approvals, paperwork etc.  Connecting what I do to the underlying reason why I do it ensures I stay energised.

One of my colleagues tells me I have a look in my eye that tips them off I am about to ask "Why .... ?"  That person has learned to be prepared for the fact I may ask a series of questions to try to connect what they asking me for/about to a wider context.  That helps make sure everyone is on the same page and that we are taking action that supports a common purpose (or conversely identifies that taking that a certain action doesn't support that purpose and won't be pursued).

If you're a student looking for an opportunity after graduation (employment, or postgraduate study) I'd encourage you to make sure you not only communicate what you can do, but why using those skills in a particular organisation excites you and aligns with what makes you tick.  If you can join those dots I believe there is a better chance that doors will open.

For anyone who wants to think more about their own "why" a one good place to start may be this page from Forbes.  If you can stay connected to why you do what you do then 2017 will be all the more exciting!





My thoughts this week start from an email a student who has just finished a conjoint degree sent me recently.  The subject line was "Thank you for 5 years of awesome!"  She'd really enjoyed the time she had spent in the Department and acknowledged the culture in the Department that aims to ensure students have an outstanding experience while they are with us.   In turn working with those students means the roles our staff have can offer a lot of job satisfaction.

Thinking about 2017 I want to explore how to use the skills, knowledge and energy of staff and students here to help build New Zealand into a country that is more "awesome" for all concerned.  Once we return from a summer break we'll be moving into a round of bidding for government research funding through the Ministry of Business, Innovation and Employment (MBIE).  The Ministry stated an aim in 2015 "to support an increase in real median household income of 40 per cent by 2025, from $1,300 (in 2012) to $1,800 per week".

Supporting that target MBIE have three objectives:

  • more competitive businesses – doubling labour productivity growth and increasing the real exports to gross domestic product (GDP) ratio to 40 per cent
  • job opportunities for all – achieving an ongoing unemployment rate below 4 per cent
  • affordable housing – a lower ratio of housing cost to income.

MBIE acknowedges these objectives are ambitious, and they challenge any organisation seeking research funding to provide a program of work that supports these objectives.

An example of work being done in Engineering Science to address these challenges is our participation in the Science for Technological Innovation National Science Challenge.  The Department is engaged in Portfolio 4 of the challenge which "aims to enable organisations to combine analytics techniques and new ICT methods to create value from large quantities of data through better processing, presentation, collaboration and decision-making tools."   Portfolio 4 is led by Professor Andy Philpott.  It has a spearhead project known as R five which is "a collaborative modelling effort looking at the effects on New Zealand businesses of randomness, risk, rivals, remoteness and resource limitations."  The project includes a particular focus on Māori capacity development (led in the Department by Assoc. Professor Andrew Mason).

The Faculty of Engineering has recently signed a new partnership with the Federation of Māori Authorities and looks forward to building closer ties between engineering researchers and Māori enterprises.

So with MBIE bids firmly on my list of "homework" for 2017 I will sign off for 2016.

Best wishes for Christmas and the New Year,


The Department of Engineering Science hosts undergraduate degrees in both Engineering Science and Biomedical Engineering.    I enjoy seeing examples of how the natural world can offer solutions to traditional engineering problems.  Much of my own work is in geothermal energy and one example in that industry is the bioreactor at Contact Energy's Wairakei plant that removes hydrogen sulphide from cooling water used in the plant, before the water is discharged to the Waikato river.  In the bioreactor the cooling water flows through 378km of pipes in which sulphur-oxidising bacteria floursh and remove 80% of the hydrogen sulphide.  The bioreactor is a world first, and was New Zealand Energy Project of the Year in the Deloitte Energy Awards.


At the University I enjoy seeing the developments that come out of the Auckland Biomimetics Lab, hosted in the Auckland Bioengineering Institute, and run by Associate Professor Iain Anderson.  The work done in the lab involves

"drawing inspiration from nature to develop new technologies. Living organisms and natural phenomena have certain behaviours and properties which let them exist in harmony with the surrounding environment. By understanding these natural processes, we are developing technologies to venture into new territories."

One of the lab's projects is building a human powered racing submarine (known as "Project Taniwha") which they race in an international contest each year.  This year the team won the overall trophy, as well as awards for being the most reliable submarine and the best non-propeller powered submarine.  The team are the only Southern Hemisphere team.  Curious how such a vehicle works?  It's pedal-powered, with a diver inside who breathes from a dive tank.

“We also got the reliability trophy because we didn’t abort once in 18 runs around the circuit,” says Iain. “That’s about five kilometres underwater. New Zealand has proven to the world that Taniwha is a top sub with a top team.”

“They were very impressed with our innovative bending tail on the sub that takes the place of a rudder,” says Iain. “The rear section of the Taniwha bends back and forth like a fish and enables the Taniwha to turn without a rudder.”

The Taniwha is a great mix of creativity, determination, and Kiwi enginuity - with some inspiration from nature.  If you're visiting our Department (at Uniservices House, 70 Symonds St) you will see it on display on the ground floor level.  I know there's not a huge market for human power submarines ... however the Biomimetics Lab also applies it's expertise to a range of other ventures with commercial impact, including a spin-out company called Stretchsense which develops wearable technologies.


New Zealand has the dubious honour of having the highest rate of teen suicide in the developed world.  As a New Zealander I am not proud of that.  Overall the number of New Zealanders who take their own lives is normally larger than the number who die in road accidents.  As a community we discuss the road toll freely, but choose not to discuss mental health as openly.  I want to join others who challenge this taboo.

I am very mindful of the stresses and mental health difficulties all University students face. The young people in the Department of Engineering Science are not immune from the difficulties of navigating financial struggles, chronic health problems, bereavement, relationship difficulties, questions of identity and more.  Those challenges seem to be increasing - as is my awareness of the potential for incidents of self-harm.  I have an open door policy with students and over the years have had students disclose some fairly mind-boggling circumstances - naturally confidentiality means I will not expand on the details.   We're one of the smaller Departments in our Faculty and have a strong sense of community.  We aim to proactively check-in with students if we see any signs of mental health issues that go beyond the realms of normal student life. I am grateful for the support of Faculty's student engagement team who help support students in a range of ways.   The University of Auckland also has an excellent health and counselling service.
Mental health

Engineers are problem solvers - and I am pleased that two staff I work with have devoted some energy to the problem of youth mental health.  Dr Mike O'Sullivan Jnr and Associate Professor Cameron Walker have been working with the Adolescent Health Research Group to help analyse data from the Youth 2000 series of surveys.  Recent work conducted a final year research project by an undergraduate student (Jung Shaan Lee) working with Mike and Cameron used modern statistical tools on the Youth 2000 data set and identified three key themes associated with depression in young people - having poor family relationships, feeling unsafe or out of place at school, and experiencing difficulty accessing mental healthcare.  Jung Shaan's work helped shed light on the interconnectedness between multiple factors and has already attracted the interest of mental health professionals working in the Adolescent Health Research Group.  I think that's a great use of some engineering expertise.


In an emergency situation (in New Zealand) please call the Mental Health Crisis Service on (0800) 800 717. This is a free public health service. The crisis teams are part of the Community Mental Health Service and are available 24 hours a day, 7 days a week.

For free mental health counselling please call Lifeline on (0800) 543 354, 24 hours a day.