I learned structured decision-making (SDM) or multicriteria analysis (MCA) during my PhD and have published several papers on the topic. When it comes to the tough decision of buying a hippo house, I just cannot help using the approach in my own life ;).
Step 1: I interviewed the other stakeholder and decision-maker, Mr Hippo and asked his criteria for evaluating a potential house to buy.
It didn't take him long to give me the three things on his mind: value for money, location and niceness of the house. Well, this is a great start but these are rather fuzzy concepts, so we moved to the next step.
Step 2: We had a brainstorming session to flesh out what exactly on Mr Hippo's mind. The figure below, in the form of a mind map, shows the result of the session. Now we got four criteria (cost, location, restriction on renovations, and quality) and 30 plus sub-criteria.
No house will probably meets all the 30+ requirement (we are too lazy and architecture-challenged to build our own) and some trade-offs have to be made.
Step 3: We then ask ourselves what are the deal-breakers from the list of 30+ subcriteria. Those in brown are Mr Hippo's top concerns, and the green are mine. The ones in red are the common deal-breakers.
Step 4: I asked the question of "why is it so important?" to try to find the root cause for our concerns. This step is VERY important because some seemingly unrelated concerns are in fact due to the same underlying reason. If we include all concerns but not their root cause in the decision-making process, we will commit the crime of double counting. For example, I want a well-functioning kitchen and Mr Hippo refuses to look for houses that are "irredeemably ugly." Kitchen's functionality and a house's ugliness seem to be very different matters, but they both link to the same root cause: the potential cost of renovation. Neither of us is ready for spending a fortune on renovating a house, so I want good kitchen appliances that are ready for me to cook a storm with, and Mr Hippo doesn't feel like conducting a project to beautify our house from its roof to toe. House-hunting itself seems to be lots of work already...
Anyway, here is the set of seven criteria we settled on:
1. Reducing the cost for house purchase,
2. Reducing the cost for house renovation,
3. Decreasing time spent on traveling to work,
4. Decreasing noise from traffic,
5. Increasing space for recreation,
6. Increasing living space, and
7. Increasing mental health.
Step 5: I then found an indicator for each of the seven criteria. The selection of the indicators is also very important, and for our purpose, we need something that are heuristic and easily accessible. Remember we only got on average 5-10 minutes to look for a house, and we need to find information to assess each indicator during this short of period. So here are the corresponding indicators and their units:
1. Price of the house ($),
2. Can we rebuild kitchen (Yes/No),
3. Time to travel to Mr Hippo's work (minutes),
4. Level of noise from traffic (low, medium, and high),
5. Size of garden (m2),
6. Is main bedroom big enough for a king-sized bed? (Yes/No), and
7. Is bathtub big enough for Mr Hippo? (Yes/No).
Step 6: To collect information for the seven indicators for each house. I have put together a table with the indicator in one dimension and the address for each house we are going to see tomorrow in the other. The idea is to focus on the things we really care about (the seven criteria) and avoid being distracted during house-seeing ("Um, that carpet is irredeemably ugly..."). I am confident that we could find all the information within 5 minutes (information for a couple indicators was gathered beforehand, including house price and size of the main bedroom).
I look forward to testing the effectiveness of my SDM approach tomorrow, and of course, this is going to be iterative process and we will updating our criteria and indicators as we trample new tracks! I'll keep blogging about anyprogress, so stay tuned!
P.S. SDM/MCA invovles many more steps than what I documented here.
No comments:
Post a Comment