Parameter | Environment attribute definition and values | Data and empirical foundation |
---|---|---|
\({sl}_{m}\) | Slope penalty \(\left\{{sl}_{m}\in {\mathbb{R}}\,|\ {sl}_{m}>0\right\}\) | The slope penalty was applied to agents traveling from peripheral areas of the city to/ from the central business district (CBD) since these trips involve at least one journey negotiating an uphill slope (0.5%). The use of this penalty aligns with the empirical literature which found that the attractiveness of both walking and bicycling decreases with increasing slope [34, 35]. Remains constant For walking and bicycle trips from peripheral areas of the city to/from CBD slope penalty \({sl}_{w}=1.05\); \({sl}_{b}=1.125\). For all other walking and bicycle trips, no penalty is applied i.e., \({sl}_{m}=1\). See Supplement S2d for more information |
\({wt}_{im}\) | Waiting time distributions, for bus and BRT, by SES \(\left\{{wt}_{im}\in {\mathbb{R}}\,|\ {wt}_{im}>0\right\}\) | Bus and BRT waiting times (in minutes) by SES, assumed to be normally distributed with mean (\(\widetilde{x}\)), sd, maximum and minimum (set to zero) informed by data derived from the 2019 Bogotá Household Travel Survey [27]. Agents draw a new value each time step Bus: SES1 [\(\widetilde{x}\) =18.6, sd = 13.6, max = 90]; SES2 [\(\widetilde{x}\) =15.1, sd = 11.9, max = 127]; SES3 [\(\widetilde{x}\) =14.8, sd = 12.1, max = 120]; SES4 [\(\widetilde{x}\) =12.0, sd = 8.8, max = 60]; SES5 [\(\widetilde{x}\) =11.6, sd = 9.7, max = 40]; SES6 [\(\widetilde{x}\) =11.9, sd = 8.4, max = 40] |
BRT: SES1 [\(\widetilde{x}\) =16.5, sd = 10.9, max = 60]; SES2 [\(\widetilde{x}\) =14.5, sd = 10.6, max = 75]; SES3 [\(\widetilde{x}\) =12.5, sd = 9.3, max = 90]; SES4 [\(\widetilde{x}\) =11.0, sd = 7.5, max = 45]; SES5 [\(\widetilde{x}\) =9.9, sd = 5.9, max = 30]; SES6 [\(\widetilde{x}\)=8.6, sd = 5.4, max = 20] All waiting times are expressed in minutes | ||
\({s}_{im}\) | Mode speeds \(\left\{{s}_{im}\in {\mathbb{R}}\,|\ {s}_{im}>0\right\}\) | Average car [36], bus [37], BRT [38], bike [39] and walking [40, 41] speeds informed by the literature and converted to meters per minute in the model. Given the absence of information concerning motorcycle speeds in Bogotá, we assumed motorcycles travel slightly faster than cars given their ability to weave between cars Car = 35 km/h (583 m/min); Motorcycle = 40 km/h (667 m/min); Bus = 13.7 km/h (228 m/min); BRT = 26 km/h (433 m/min); walking = 4.8 km/h (80 m/min); bicycle = 17 km/h (283 m/min) |
\({C}_{m}\) | Mode cost \(\left\{{C}_{m}\in {\mathbb{R}}\,|\ {C}_{m}>0\right\}\) | Mode costs include fuel costs [42] (for car and motorbike, which are calculated based on the total travel distance), fare costs for bus and BRT (full price and subsidized fares for those meeting criteria for SISBEN (Sistema de Selección de Beneficarios) subsidy ~ 10.2% of population, excluding older people and people with disabilities (assumed to be poorest 10.2% of agents)) [43, 44], and parking costs [45] (for bike, car and motorbike, which depend on level of parking demand in agent’s work zone, and level of parking service—parking at a facility, on a concrete floor or on a grass floor—which is randomly chosen each day) Fuel costs: motorcycle & car = 0.23858 COP/m Fare costs: a) Full price fares—BRT = 2400COP; bus = 2200COP/ trip; b) SISBEN subsidized fares—BRT = 1991COP; bus = 1825/trip (average per cost estimated by assuming 260 business days/year ~ 22 days/month or 44 return trips, up to 30 trips at SISBEN subsidized price (BRT = 1800COP; bus = 1650COP [43]) and remaining trips at full price) Parking costs: bike = 4800COP/day; motorcycle: high demand \(\in [17760 \,24960 \,35520]\) & low demand \(\in [14400 \,19680 \,28320]\) facilities (COP/day) Car: high demand \(\in [25440 \,35520 \,50400]\) & low demand \(\in [20160 \,28320 \,40320]\) facilities (COP/day) |
\(Pmpp\) | Percentage of motorcyclists paying for parking \(\left\{Pppm\in {\mathbb{R}}\,|\ Pppm>0\right\}\) | The percentage of motorcyclists paying for parking in the model is 55%, as determined though model calibration. We assumed that only a fraction of motorcyclists use paid parking facilities given that illegal parking, including frontage parking – parking on sidewalks and spaces between the buildings and the street – is not uncommon |
\({RC}_{mg}\) | Rate of personal crime on mode \(m\) for agent with gender \(g\) \(\left\{{RC}_{mg}\in {\mathbb{R}}\,|\ {0<RC}_{mg}<1\right\}\) | The prevalence of personal crime was informed by 2015 crime [46,47,48,49,50] and trip data [51]. Crime data were not available for bicycle, car and motorcycle trips so we made assumptions about the crime prevalence relative to the other modes. Furthermore, we apportioned these overall crime prevalence rates to men and women, based on data from mobility surveys [51]. (See Supplement S2e for more information). Crime prevalence rates by mode remain constant Car: male = 0.007% & female = 0.008%; motorcycle: male = 0.007% & female = 0.003%; BRT: male = 0.027% & female = 0.048%; Bus: male = 0.011% & female = 0.017%; bike: male = 0.370% & female = 0.130%; walking: male = 0.328% & female = 0.590% |
\({n}_{mz}\) | Number of bus and BRT stops, by SES stratum \(z\) \(\left\{{n}_{mz}\in {\mathbb{R}}\,|\ {n}_{mz}>0\right\}\) | Geospatial data were used to derive the number of bus [52] and BRT [53] stops in each SES stratum (see Supplement S2e for more information). For simplicity, bus stops were randomly distributed within each SES stratum, while BRT stops were roughly distributed according to BRT route maps Bus stop count: SES1 = 620; SES2 = 2,323; SES3 = 2,835; SES4 = 769; SES5 = 327; SES6 = 236; Total = 7,110 BRT stop count: SES1 = 0; SES2 = 12; SES3 = 76; SES4 = 27; SES5 = 8; SES6 = 3; Total = 126 |
\({n}_{wz}\) | Number of workplaces \(w\), by SES stratum \(z\) and within the CBD \(\left\{{n}_{wz}\in {\mathbb{R}}\,|\ {n}_{wz}>0\right\}\) | Estimated using data from the empirical literature [31]. For simplicity workplaces were randomly distributed within each SES stratum and the central business district (see Supplement S2c for more information) Workplace count: CBD = 16,495; SES1 & 2 = 3,705; SES3 = 11,351; SES4, 5 & 6 = 15,188 |