CHAPTER 3 ESTIMATING WALKING DISTANCE
3.2 Energy expenditure experiment
For the following energy expenditure analyses, the predictive function
x needs to be established. Thus this research designed an experiment and would recruit 30 participants, 15 men and 15 women, to conduct experiments on seven distinct types of streets. In order to determine the effect of timing (day or evening) and to enlarge the sample size, the subjects would be asked to walk up and down each street, once during the daytime and once in the evening (after dark). Thus, a total of 840 samples would be collected to perform a regression analysis. It was a compromise necessitated by the limited budget.19
Streets of seven main types in Hsinchu City, Taiwan would be investigated in this experiment, including arterial roads, boulevards, residential streets, mixed-use streets, commercial parkways, downtown streets, and alleys (Marshall, 2005). The locations of those streets are shown in Figure 6.
Figure 6 These seven experiment sites were chosen to represent the main street types in Taiwan.
The street spaces are shown in Table 3.1. There are five street elements representing the attributes of these streets in Table 3, including right of way (ROW), width (WD), planting (PL), lighting (LG) and land use.
As to the instrument, this research would use a Polar RS300Xsd heart rate monitor consisting of a watch-style computer, a chest strap, and a speed-distance transmitter to record the wearer’s energy expenditure. Before testing, the Polar RS300Xsd asks for the participant’s characteristics (birthday, body weight, height and gender) to be inputted, and to test the participant’s fitness to determine the max. oxygen consumption (VO2max) and max.
heart rate ( fH ), which are then used as the individual parameters. VO2max is the maximum capacity of the body to contain and utilize oxygen during incremental exercises.
This capacity reflects the physical fitness of the individual and accounts for a significant portion of the variation in metabolic rate (Poehlman et al., 1990). The device also records the walking speed, walking time, and walking distance, all of which are required to analyze the terrain factor value through equation (8).
20
Table 2 Spatial characteristics of the experiment sites.
Experiment
Land use Planting Lighting (Lux)
1 Arterial road Shoulder 2 Shared Mixed-use Shrubs 6
2 Boulevard Sidewalk 5 Exclusive Mixed-use Trees 8
3 Residential
street
Sidewalk 2.5 Exclusive Residential Trees 6
4 Commercial
Arcade 1.5 Exclusive Commercial Potted plants
14
7 Alley Lane 4 Shared Residential Potted
plants 4
Note: Width, right of way, land use, planting and lighting are independent variables of regression analysis. Their codes are shown in Table 5.
Prior to the tests, each participant would be informed as to the purpose of the experiment and received training in the operation of the Polar RS300Xsd. Then the subjects tested their personal VO2max value as the parameter for the device to compute the WEE while avoiding any distractions. Heavy physical exertion, smoking and consuming a large meal prior to the test are also avoided. The participants refrained from frequently visiting those streets prior to the test in order to avoid becoming too familiar with those sites, which could then result in a lower stimulus.
During walking, the participants would be asked to maintain as much possible a constant walking speed in order to reduce the inaccuracy resulting from using the mean walking speed as the input velocity v. In addition, the subjects would be asked to rest for ten minutes before their next test to avoid the effect of the previous walk influencing their metabolic rate in the next test. All experiments would be carried out in dry weather.
21
Table 3 Summary of independent variables.
Variable Data Scale Value (Unit) Description Physiological factors
Gender (GE) categorical Male=1 Female=0
To reflect the physiological difference between male and female
Right of way(ROW) ordinal Shared=1 Exclusive=0
To indicate pedestrians walking under disturbance or not.
Width(WD) continuous (m) Degree of space offering pedestrians
unobstructed walking Planting(PL) ordinal Potted plants=1
Shrubs=2 Trees=3
The level of street planting coded as a ordinal variable indicating more green area, more pleasant surrounding
Lighting(LG) continuous (Lux) Facility to indicate a safe walking environment after dark
Land use factors
Commercial(CM) categorical Commercial=1 Mixed-use=0
Land use with various commercial activities as the main attraction for the pedestrians
Mixed- use(MU) categorical Commercial=0 Mixed-use=1
A popular pattern of land use in Taiwan mixing retail outlets, residential, service and other functions
Residential(RS) － Commercial=0
Mixed-use=0
Land use with low intensity and simple function for residential use only
Environment factors
Evening(NI) categorical Evening=1 Day=0
To differentiate walking in daytime or evening
Temperature continuous (℃) A degree of comfort in the open air; the range is from 18℃ to 28℃
Note: Land use is indicated by means of dummy variable, thus RS is absence of the independent variables.
22
The questionnaire consists of basic personal data, origin-destination pair survey, and the street characteristics along the routes. For the personal basic data, each respondent would be asked to provide personal basic data including age, gender, body weight (kg), and the load they carried (kg). The load carried would be recorded by the respondent checking the six enumerating types of carried items marked with weight, such asanotebook (2kg). Each of the respondents described their daily commuting trip using an origin-destination pair table, as shown in Table 4.
Table 4 Example of O-D pair table.
Links of trip
Origin Destination Mode of transportation
Walking distance(m)
Duration(min) Street type
Slope(%)
Ⅰ A B 1 200 5 5 0
Ⅱ B C 2 － 10 － －
Ⅲ C D 3 － 13 － －
Ⅳ D E 1 450 8 7 0
Note: Mode of transportation, 1. walking, 2. bus, 3. rapid transit, 4. bicycle, 5. motorcycle.
Each respondent then identify all places they visited during their trip on a map (such as origin, transit stop, and destination) and then link them as a route. A commuting route might involve several links divided by mode of transport or street type, but the walking link is the only item of concern in this research. The participants would be asked to just write down the walking time for each link, and then I measure each walking distance. Walking velocity would be obtained by dividing each walking distance by the walking time. To compute the adjusted terrain factor value through equation (8), each respondent would mark down how they felt about the right of way, lighting, and planting (see Table 3) by check-marking the items which are shown complete with a sample photograph. For example, the respondents indicate the lighting level from a picture, and then the corresponding Lux number would be used as the input value.
It is unrealistic to assume that respondents can or will spend time on measuring VO2max through a device during the interview. Fortunately, Shvartz and Reibold(1990) provides an age-based method to check the VO2max value after the participant provides the information on their age and gender and how long s/he participates on a regular basis in sports per week.
23
This allows us to apply equation (8) to calculate the estimated metabolic rate and multiply it with the walking time in each street type. I then summed them using equation (4), and then estimate the walking energy expenditure for each commuting trip:
K
k
k k
w w t
e
1 (9) where K denotes the number of street types; w denotes the metabolic rate while walking k on a type k street, and t denotes the walking time on a type k street. k