Especially true for city condominiums, being able to walk to nearby amenities and transportation choices is becoming more important by the day (and especially so with the horrendous KL traffic which doesn’t seem to get better). So for the past two months, our small R&D team here at GoodPlace.my have been currently experimenting with a new feature that computes the walkability of a neighbourhood by pulling together a couple of data points to compute what we call the GoodPlace Walkability Scores (GWS).
The GWS is built on top of Google Maps API and is scored using the walking distances to:-
- Basic amenities such as restaurants, clinics and schools.
- Secondary amenities such as drinking spots, entertainment outlets, laundrettes, shopping centres and police stations.
- Transportation points (taxis, train stations).
We apply a “dampening factor” on the overall score which increases logarithmically with distance. The maximum GWS score is 100.
We have debuted this feature at our sister site, the KLCC Condominiums Database, and you can download a map with the assigned GoodPlace Walkability Scores here. We are currently working on maps for Mont Kiara and Bangsar – watch this space.
Credits: This feature was suggested by EC and Harry – two of our loyal readers at http://klcccondominiums.com.my/. We also cannot claim authorship over the concept; in fact we have liberally borrowed ideas from Walk Score (http://walkscore.com) although our algorithm is independently derived and proprietary.
Addendum: The Algorithm
The maximum Walkability score is 100. The initial point is therefore set at 100 which then gets subtracted through a series of weighted scores which each represents a particular facility. Mathematically, this can be encapsulated in this simple formula –
The heart of the algorithm is the computation of each of these weighted scores represented by X in the equation above. Each score again starts at a maximum value (which varies between the type of facility which is correlated to the degree of importance) which gets “dampened” further by two aspects:-
- Distance between the property and the facility
- “Ease of walk”: existence of pathways and walking lanes
The data points above are derived from Google Maps API (documentation here). The ease-of-walk factor needs a little more processing compared to the walking distance which is comparatively straightforward.
In the equation above, A represents the coefficient which represents the degree of the importance of the facility. Typically, the coefficients for safety related facilities (guard houses, police stations) and education (schools and colleges) are higher than, say, the laundrette and pizza eateries. W is the ease-of-walk factor while d is the path distance (i.e. not the straight line distance) between the facility and the property.
For each type of the facility in consideration, we typically build matrices since there will be multiple locations for each facility (for example, different transportation points for taxis and buses) –
Each permutation of these are reduced and factored into a single score which then fed back into the main equation to derive the final walkability score.
With new facilities and walking pathways being built, obviously the scores change, and we try to do a new what we internally call the Crawl-And-Compute process once every six months. For latest scores for existing properties inside our database, contact me.