Proud of my Masters Program in OR Winning the George D Smith Prize

This year's INFORMS Business Analytics and Operations Research conference was held in Huntington Beach, California. Excellent talks on really relevant topics. Great conference, networking and top organizers. :)

Best of all, my former masters program in Operations Research, the Centre for Operations Excellence at the Sauder School of Business, University of British Columbia, won this year's George D Smith Prize. This is the field's top prize recognizing some of the best education programs.

The UPS George D. Smith Prize is awarded to an academic department or program for effective and innovative preparation of students to be good practitioners of operations research.

I was honoured to be at the gala event, and ecstatic for the program's win. My career would not have looked quite the same without the program. So happy for them.

"I am indebted to my father for living, but to my teacher for living well." - Alexander the Great

"There are two kinds of teachers: the kind that fill you with so much quail shot that you can't move, and the kind that just gives you a little prod behind and you jump to the skies." - Robert Frost

What I learned from a sabbatical year

I spent 2013 'overlanding' through South America with my partner. 1 year, 1 continent, 1 simple car, 2 people, 13 countries, 40,000 km. After moving from Canada to the UK 5 years ago, and setting up a new life there, we gave up our jobs, salary, friends and all the comforts of life in one of the greatest metropolis in the world. A lot to let go, but we gained so much more.

Above all, I learned how little I need to live on to be happy, material-wise. We converted the back of our little van into a bed, so we slept in it a lot of the nights. Wild camping at some bizarre and cool spots, like 24-hour gas stations and garages, road-side somewhere in the country, cliff edge by the sea, a lot of central plazas and town squares, in front of police stations (with permission), and once within a secure military compound. The living was rough, and it took some getting used to. I had very few possessions; I was happy; and my eyes were filled with wondrous things throughout the year.

Communities kick ass in supporting overlanding travellers of all modes, by car, motorcycle, bicycle, uni-cycle or even by donkey(!). A few hundred people gathered on a Facebook group were the best near real-time information providers. Almost all overlanders are super eager to share information with each other and help, because we've all known a few hard moments on the road. Most people have never met each other in the cyber community, but are ready to answer questions when asked.

One can have too much of a good thing. I love travelling, and still do. 60+ countries later, my imaginary list is still quite long. Doing a year of pure travel is super fortunate, and I almost don't dare to utter that sometimes I found it hard to drag myself for the 11th time in 3 months to drive through yet another beautiful wine country with breath-taking alpine scenery, or more Andean mountain villages, or serene beaches... etc.. Managing the trip is a huge challenge, but I also missed work a lot, missing the other challenges. So, in the evenings I:

• brushed up on R and some Machine Learning techniques through Coursera (awesome!)
• learned something new, Octave, Python, more Machine Learning techniques
• read a lot of blogs on OR, analytics and data science
• wrote a few blog articles here (definitely neglected when I was working a busy job)
• thought long and hard about what I want to do when I get back

A bunch of random stuff I learned a bit about:

• Navigating in places I've never been before. "Don't listen to the British lady (aka the GPS voice), she's never been to Venezuela", and she's leading us down a dead-end.
• Spotting and dodging potholes, rocks, livestock, cowboys, donkey carts, tree stumps, burned tires (12 day riot aftermath), flying fallen ladder (kid you not from the truck 15m in front at 90km/hr), alignment-breaking and bottom-scraping grooves in the road from heavy Brazilian trucks ... ...
• Making it Swimming through potholes the size of a swimming pool, with muddy and seemingly endless bottoms, with a 2x4 car that had 6" clearance (nope, didn't get stuck even once! 4x4 is not a necessity for everyone)
• Fixing cars and dealing with mechanics, and their other-worldly Spanish
• Playing with the police to always avoid paying bribes (wasn't too often)
• Finding out just how friendly people are (lots of home-stay invites)
• Playing the Quena (Andean flute) is way harder than it seems - sticking with my uke instead
• Optimising the journey in Travelling Sales Man fashion (had to return to the origin to sell the car) - yes, Operations Research is useful in every walk, or drive, of life
• Optimising decisions under uncertain conditions
• And of course, learning Spanish, with all sorts of accents and idioms, and the 13 countries' history, culture, landscape, food and people (P.S. mechanics and old country farmers are really hard to understand)

Having finished the year-long journey over a month ago, I was inspired to write this article after reading "Why I put my company on a year-long sabbatical". This is not a PR article, but one to say that anyone can do this sabbatical thing, and you will learn a ton. You don't need the best car all decked out. You don't need to be young. You don't need to be retired. You don't need to be without kids (met a lot of families, with kids from 6-months to 17-year olds). You don't need to have a partner. You don't need to be rich (our all-in costs: £10,000 per person, assuming 2 people sharing). Actually, you will learn how little you need at all. All you need is a bit of discipline to save some money, a bit of gut to throw yourself at it, some luck and common sense to be safe, and a lot of curiosity to explore.

In case you are inspired to consider a sabbatical year, here are some great overlanding resources:

Expedition Portal

WikiOverland

LandCruising Adventure

2014 is going to be great. I am never more ready.
First step, land an awesome job.

More MOOC on Analytics - Coursera

A hoard of analytics related Massive Open Online Courses (MOOCs) are about to start in September. Have your pick on what to learn. Having taken a few Coursera courses now, I would recommend 1) not taking too many courses at once, however tempting it is to sign up to all of them, unless you have no other work or projects on the go. This is just to make sure you have a reasonable load and able to devote enough of your attention to learning the material properly. 2) Make good use of the discussion forums, as they are both a good source of clarifications and a window into other people's perspectives on the material. 3) Do the exercises, programming assignments and quizzes to ensure your understanding of the material.

Linear and Integer Programming
Starts 2 Sept 2013, 9 weeks, 5-7 hours/week
(the basics of mathematical optimisation, a core toolkit in the field of Operations Research)

Statistics One
Starts 22 Sept 2013, 12 weeks, 5-8 hours/week

Introduction to Recommender Systems
Starts 3 Sept 2013, 14 weeks, 4-10 hours/week

Computing for Data Analysis
Starts 23 Sept 2013, 4 weeks, 3-5 hours/week
As I've written before here.

Web Intelligence and Big Data
Starts 26 Aug 2013, 12 weeks, 3-4 hours/week

Thinking Again: How to Reason and Argue
Starts 26 Aug 2013, 12 weeks, 5-6 hours/week
Perhaps a bit off topic, but perhaps not, since all analytics are more or less rooted in proving or disproving arguments, so we better learn how to do it well.


Related article:
Coursera and the Analytics Talent Gap
Starting up in Operational Research: What Programming Languages Should I Learn?

Learn R with Coursera for Data Analysis

Heads up: the Computing for Data Analysis course is running in September 2013.

It will teach you the R language for data analysis. The course is described as:

This course is about learning the fundamental computing skills necessary for effective data analysis. You will learn to program in R and to use R for reading data, writing functions, making informative graphs, and applying modern statistical methods. 

In this course you will learn how to program in R and how to use R for effective data analysis. You will learn how to install and configure software necessary for a statistical programming environment, discuss generic programming language concepts as they are implemented in a high-level statistical language. The course covers practical issues in statistical computing which includes programming in R, reading data into R, creating informative data graphics, accessing R packages, creating R packages with documentation, writing R functions, debugging, and organizing and commenting R code. Topics in statistical data analysis and optimization will provide working examples.

Related article:
Coursera and the Analytics Talent Gap
Starting up in Operational Research: What Programming Languages Should I Learn?

School uniforms in developing countries: An unnecessary evil? - High-level test

Earlier I wrote a post about the requirement for school uniforms in developing countries and how I saw this as a potentially offensive injustice. I completed the first step by forming my hypothesis, "The unnecessary requirement for school uniforms in developing countries puts undue financial stress on families already struggling to afford basic necessities and/or tuition, and potentially even excludes some children from attendance." Now I am looking to test that hypothesis quickly at a high level. I want to do some research to gain reasonable assurance that the hypothesis is correct before I might move on to establish the magnitude of the problem.

Schools for Africa is a UK Registered Charity mainly focused on building schools, but who also say: "£40 will buy 10 sets of primary school uniforms". To put this into perspective:

• They also say: "£235 will buy 50 text books for the children to share". That's £4.70 per textbook vs. £4.00 per uniform.
• £4 is about the same as an average day's wages in Ghana
• $2,500 USD GDP PPP per capita de-adjusted for PPP, converted to GBP and divided by 365 is £3.02
• £4 is about the same as an average week's wages in Ethiopia
• $1,000 USD GDP PPP per capita is about £4.23 per week calculated as above, sadly reasonably assuming no days off
• I choose these countries as I visited them in 2011, but it is worth noting that Wikipedia reports school uniforms as required in Ghana

The folks at Project Ethiopia, an American 501(c)(3) have reportedly bought 1,695 school uniforms at $8 a piece. These uniforms are also said to last two years, so that's an annual cost of only $4. They make the relevant point that these uniforms are the only set of clothing for many, which would lower the additional burden of the uniform requirement on top of that for clothes. Note, however, that $8 is more than a week's wages as calculated above. Again for perspective:

• They also claim to buy over library books for $3 a piece
• They also claim to buy a years school supplies (5 exercise books, 1 pen, bar of soap) for $3

Gift Ethiopia, a UK Charity will provide an Ethiopian school uniform for £8, describing it as such:

Without a uniform, many children in Ethiopia are unable to attend school. Many families, especially larger ones, struggle to provide a uniform for all their children. These children are denied an education and the chance to socialise with children their own age. Your gift will provide a student with a brand new, full school uniform, ensuring they can take their place in the classroom with pride.

• £8 for a school uniform is about the same as they say it will cost to provide a school dinner for over 10 weeks

The first program listed on the website for Common Threadz, a 501(c)(3) American non-profit, is "School Uniforms for Orphans & Vulnerable Children". They describe the problem:

For families facing the challenges of poverty in Africa, school clothes are not as crucial as the next meal. The direct costs of education, from a uniform and shoes to books and stationery, force millions of orphans and vulnerable children to miss out on school each year. For a child in need from a poor rural family who may only own one pair of old pants or a tattered dress, a school uniform is not just a requirement, but essential to build confidence and academic success.

World Vision UK runs MustHaveGifts, and sells a pretty smart looking Pakistani school uniform for £12.

• The uniform is described thusly:
Pakistan: Children who can't afford a compulsory school uniform can be denied the right to an education, leaving them vulnerable to exploitation. With a school uniform, children can attend school for the very first time and get on the path to a brighter future.
• At $2,500 USD per capita PPP GDP de-adjusted to remove PPP is $941 or £1.65 per day or almost £12 per week

Based on the above I think that we can conclude that there is reasonable evidence to suggest that in parts of the developing world school uniforms are comparatively expensive and a prerequisite to education.

The next step, though I may not endeavour to take it due to the scale of effort required, is to gather all of the available evidence together to establish a high-level estimate of the scale of the problem. What is the aggregate cost of school uniforms across the developing world? How many children are denied an education as a consequence of their family not being able to afford school uniforms? Ultimately building to the question, What if the requirement were abolished? Once we know the "size of the prize", and please do forgive me for that blatant consultant-ism, we can begin sizing up what can be done about it.

School uniforms in developing countries: An unnecessary evil? - Hypothesis

There are charities helping families in developing countries to buy school uniforms for their children so that they can attend school. This is a good thing, right? Which part? The part about charities helping families in developing countries or the part where this is even a problem? If what I consider to be an arbitrary policy is preventing impoverished children from getting a primary education, this is a great injustice.

Testing this with a few friends, I have concluded that this quite possibly is the case, and I also received some stark warnings about the social, cultural, and psychological dimensions to school uniforms. These warnings are certainly valid, but many great in justices in this world have been toppled that were held up by social, cultural, and psychological factors. The question is, how big is the problem, how big are the barriers, and are our efforts best placed elsewhere?

It occurred to me that this is an opportunity to try out some strategic modelling and analysis, something that I do often in my current work. I have already completed the first step of forming a hypothesis and testing with a few peers. To pursue the problem further I would take the following steps:

1. Form a hypothesis:
   
   The unnecessary requirement for school uniforms in developing countries puts undue financial stress on families already struggling to afford basic necessities and/or tuition, and potentially even excludes some children from attendance.
2. Test hypothesis at a high level
   
   Gather whatever evidence is at hand or easily available to sense-check and/or refine the hypothesis. Might the hypothesis be true? Is it likely enough to be true enough to warrant further investigation?
3. Estimate the magnitude of the problem/scale of the potential benefits from taking action
   
   This will be much like a top-down strategic business case. The key focus will be "What if we could achieve a change?" without yet talking specifically about what actions would be required. Like the previous step, this is another gate we have to pass where we must be certain it is worthwhile proceeding. The output can also be an important number socially, as $x million lost per year or y thousand children excluded from primary education worldwide can be a useful catalyst for change as it is shared and repeated.
4. Develop a portfolio of initiatives
   
   Preferably in a brainstorming/facilitated workshop environment, work with stakeholders and subject matter experts to generate potential initiatives or interventions to address the problem.
5. Prioritize initiatives
   
   Estimate costs, benefits, and risks of each initiative and then build an action plan, selecting the highest benefit set of activities that fit within your budget or capacity while managing/minimizing risk. This is a classic Operations Research portfolio optimization knapsack problem, though in practice, problem sizes are small mathematics are rarely used.