Quantifying Transit Comfort Objectively

Research Student: Simon Li

Comfort has always been a purely subjective perception. Nonetheless, the quantification of comfort is necessary in some scenarios. One obvious case is the modelling of mode shares in transportation. It has often been cited that one main variable that differs between transit modes is comfort, although a meaningful quantification of it is yet to be seen.

A certain case study within Vancouver, BC has presented the opportunity to assess the value that is derived from having a seat on a bus. At the 99 B-Line Express Bus stop on Broadway and Commercial Drive, a certain phenomena occurs that allows for this study. In its normal layout format, two separate queues at the bus stop occur. One queue is for those that are willing to stand on the bus; the other is for those who will wait for the subsequent busses in order to obtain a seat. In this scenario, there is a clear trade-off between waiting time and the comfort of having a seat on the bus. Through observations, the proportion of people that choose either of the two choices can be obtained. With the random utility theory and the probabilistic choice theory, the value of comfort derived from the seat can be quantified in terms of extra waiting time. This is the first step in the quantification of comfort.

Nevertheless, this approach is still subjective, and will still differ from person to person. Although it can be said that comfort is purely subjective and that difference between people is inherent, an objective quantification is still of interest. With an objective quantification, an objective comparison can be made between different scenarios, even when the users are from varying populations. An example is that with an objective quantification of comfort, the comfort of taking the MTR in Hong Kong can be compared against the comfort of taking the SkyTrain in Vancouver. Human factors are removed, so that factors such as greater resistance to discomfort are not included.

One of the main difficulties of objective quantification is that it is difficult to measure comfort. One approach is to measure multiple variables that may constitute comfort. Of course, the first step in finding these variables would be to do a mass survey of what variables do transit users consider influential with respect to comfort. To name a few possible ones: ambient noise, crowding level, jerkiness, smell, etc.

The main focus of this project is to attempt to measure the jerkiness of transit in an objective manner. More specifically, it will be investigating on the acceleration / deceleration profile that a transit user experiences on a particular trip. Whether this is a main constituent of transit comfort will be a separate discussion.

The first step of this assessment is to devise a way to accurately and easily obtain acceleration profiles with respect to time. The instruments proposed are modern smart phones or MP3 players which have integrated accelerometers. The accelerometers are capable of measuring acceleration in different axis,up to a few g’s in each direction. Using various apps, the values of acceleration over time are logged and saved as data. The data will basically be comprised of a timestamp (usually at a minimum of 1/30thof a second for each measurement) column, and the subsequent recorded acceleration for the 3 axes during that time step.

A pilot of this method was performed in the winter of 2010 for a separate project. In that project, 5 minutes of acceleration log was recorded for 4 separate scenarios.

1. Express Bus – Standing
2. Express Bus – Sitting
3. SkyTrain – Standing
4. SkyTrain – Sitting

The component of gravity was removed from the readings, and the acceleration for the remaining two axes was plotted on a graph. Qualitative analysis was performed on the plots, and the shape of the profile could be commented on.

Thus, this project is an extension of the pilot from the previous project. Within this project, the goal is to be able to obtain these acceleration logs for various transit routes across the city, for different times of the day. The result will be a mapping of acceleration profile of transit routes across the city both geographically and temporally. An interpretation of the data will ensue.

Interpretation of the data has not been finalized, but will deal with various traits such as maximum amplitude of acceleration over the log, shapes, frequency of oscillations, etc. One proposed method is to convert the acceleration into impulse but summing up the absolute value of the areas under the curve. Thus, one measure of comfort between two recordings is the total impulse that the user was subjected to. Another proposed method is to record the number of time that the log passes between acceleration and deceleration. This would measure the number of times where major changes in acceleration occur.

The final outcome of this project is to be able to comment on the variations in level of comfort (in this case, constituted by acceleration and deceleration) across the city and at different times of the day. Given the results, a few further studies can be performed. The first is the feasibility of installing accelerometers into busses in order to record the driving behaviour of bus drivers. If it is assumed that it is within their powers to drive either more or less comfortably to a certain degree, then training programs can be proposed to encourage uniformity in driving behaviour. Another possible study is, as mentioned before, an inter-city or inter-geographical location study of comfort on transit from various agencies.