Content uploaded by Carolien Beckx
Author content
All content in this area was uploaded by Carolien Beckx on Mar 26, 2015
Content may be subject to copyright.
ACTIVITY-BASED POLICIES TO REDUCE HUMAN
EXPOSURE TO TRAFFIC AIR POLLUTION
Carolien Beckx1, 2, Steven Broekx2, Davy Janssens1
1IMOB, Instituut voor Mobiliteit, Universiteit Hasselt – Campus Diepenbeek.
Agoralaan Gebouw D, 3590 Diepenbeek – België.
Fax: +32(0)11 23 22 17 – Email: carolien.beckx@uhasselt.be;davy.janssens@uhasselt.be
2VITO, Vlaamse Instelling voor Technologisch Onderzoek.
Boeretang 200, 2400 Mol – België.
Fax: +32(0)14 32 11 85 – Email: carolien.beckx@vito.be;steven.broekx@vito.be
Bijdrage Colloquium Vervoersplanologisch Speurwerk 2005,
24 en 25 november 2005 te Antwerpen
Content page
Samenvatting/Summary.......................................................................................................3
1. Introduction ...................................................................................................................... 4
2. Activity-based modelling..................................................................................................5
2.1 Historic evolution .........................................................................................................5
2.2 The activity-based approach.........................................................................................6
3. Methodology......................................................................................................................7
3.1 DPSIR framework ........................................................................................................7
3.2 Activity-based modelling for air quality purposes.....................................................10
3.3 Air quality impacts of activity-based policies............................................................11
4. Link to sustainable mobility? ........................................................................................ 13
5. Conclusions and further research.................................................................................15
Bibliography........................................................................................................................16
Samenvatting
Activiteiten gebaseerde maatregelen voor het reduceren van de blootstelling aan
verkeerspolluenten
De laatste jaren vindt er een trend plaats naar het gebruik van activiteiten gebaseerde
transportmodellen om het verplaatsingsgedrag van mensen te modelleren. Activiteiten
gebaseerde modellen beschouwen de verplaatsingsvraag als een afgeleide van de
activiteiten die individuen en huishoudens wensen uit te voeren. Gecombineerd met een
milieumodule heeft het gebruik van deze modellen belangrijke voordelen naar
luchtkwaliteits- en blootstellingsanalyses toe, evenals voor de evaluatie van de impact van
beleidsmaatregelen. Het onvermogen van de traditionele vierstapsmodellen om dergelijke
beleidsmaatregelen te evalueren en om noodzakelijke data voor luchtkwaliteitsanalyses te
leveren, leidde dan ook tot de ontwikkeling van een activiteiten gebaseerde aanpak voor
luchtkwaliteitsdoeleinden. Deze paper illustreert het gebruik van activiteiten gebaseerde
modellen voor het bepalen van de blootstelling aan verkeerspolluenten en legt uit hoe
dergelijke aanpak voordelen oplevert voor beleidsmakers, ook in het kader van een
duurzame mobiliteit.
Summary
Activity-based policies to reduce human exposure to traffic air pollution
For the last decade, activity-based transportation models have certainly set the standard for
modelling travel demand. Activity-based modelling treats travel demand as derived from
the activities that individuals and households need or wish to perform. In combination with
the use of environmental models this approach has important advantages for air quality and
exposure analyses and for the evaluation of air-quality benefits from policy measures. The
inability of the traditional four-step models to evaluate responses to transportation control
measures and to provide data on all the variables required for air-quality analysis has
prompted the development of an activity-based approach for air quality purposes. This
paper reports on the use of activity-based models for the assessment of human exposure to
traffic air pollution and presents the advantages of this approach for policymakers, even
within the scope of sustainable mobility.
4
1. Introduction
An efficient transport system is a vital requirement for economic development and provides
personal mobility for activities such as work, education and leisure that are key ingredients of
modern life. But transport also contributes significantly to several environmental and health
problems, like climate change, acidification and local air pollution. Numerous studies have
indicated that exposure to air pollution increases risks of developing cancer, respiratory and
allergic diseases, and aggravates the condition of people suffering from respiratory or heart
diseases (Torfs, 2004). Quantifying the impact of traffic air pollution on the public’s health has
therefore become an increasingly critical component in policy discussion.
The last decades, several transportation control measures (TCM) have been defined to
manage both emissions and exposure associated with vehicular transport. Unfortunately, the total
vehicle miles travelled and the number of trips made increased drastically, substantially offsetting
the emission reduction through advances in technology. Due to this fact one should consider the
use of other policies aimed at encouraging efficient trip chaining and scheduling of activities.
Such policy measures could have the potential to lead to significant reductions in vehicle
emissions but these assumptions need conformation. Questions that involve the linkages between
a set of travel decisions and activities can not be examined through a traditional four-step
transportation model. This kind of model focuses on individual trips where the spatial and
temporal interrelationships between all trips are ignored. The use of an activity-based model on
the contrary will put a new perspective on the research of alternative policy measures offering a
lot of advantages for air-quality purposes and policy evaluations (Recker, 1998; Shiftan, 2000).
The major idea behind activity-based models is that travel demand is derived from the activities
that individuals and households need or wish to perform, with travel decisions forming part of the
broader activity of scheduling decisions (Ettema and Timmermans, 1997).
The remainder of this paper is organized as follows. Section two gives a brief introduction
towards activity-based modelling. The use of activity-based models for air quality purposes and a
description of activity-based policies are explained into detail in section three. Next, the potential
contribution of this research to the sustainable mobility concept is described briefly. Finally, the
paper concludes and defines some topics for future research.
5
2. Activity-based modelling
Activity-based transportation models have certainly set the standard for the last decade of
modelling travel demand. This section presents briefly the historic evolution towards activity-
based modelling and describes the most important features of the activity-based approach.
2.1 Historic evolution
Modelling traffic patterns has always been a major area of concern in transportation research.
Since 1950, due to the rapid increase in car ownership and car use in the US and in Western
Europe, several models of transport mode, route choice and destination were used by
transportation planners. These models were necessary to predict travel demand on the long run
and to support investment decisions in new road infrastructure which originated from this
increased level of car use. In these days, travel was assumed to be the result of four subsequent
decisions which were modelled separately. Those models are also referred to within
transportation literature as Four-Step trip-based models (Ruiter and Ben-Akiva, 1978). A lot of
these aggregate Four-Step models failed to make accurate predictions. The major drawback
clearly is the focus on individual trips, where the spatial and temporal interrelationships between
all trips are ignored. Furthermore, the overall behaviour is represented as a range of constraints
which define transport choice, while it is in fact both an outcome of real human decision making
and of a complex choice process. The last drawback clearly is the complete negation of travel as
a demand derived from activity participation decisions.
From the mid seventies onwards the original Four-Step models were replaced by tour-
based models where multinomial logit techniques and more sophisticated techniques such as the
nested logit and probit models formed the core of transportation modelling practice (Daly et al,
1983). In the tour-based model, trips are explicitly connected in chains that start and end at the
same home or work base. By means of this property, the lack of the spatial interrelationship,
which was so apparent in the Four-Step trip based models, is dealt with. Nevertheless in these
models, travel still has an isolated existence and the question why people undertake trips is
completely neglected. This is where activity-based transportation models came into play.
6
2.2 The activity-based approach
The contributions of Hägerstrand (1970), Chapin (1977) and Fried et al. (1977) are the
undisputed intellectual roots of activity analysis. Hägerstrand suggested the time-geographic
approach characterizing a list of constraints on activity participation. Chapin has identified
patterns of behaviour across time and space. Fried, Havens and Thall have dealt with the social
structure and the question of why people participate in activities. These contributions came
together in a study of Jones et al. (1983), where activities and travel behaviour were integrated.
This was the first initial attempt to model complex travel behaviour.
The major idea behind activity-based models is that travel demand is derived from the
activities that individuals and households need or wish to perform, with travel decisions forming
part of the broader activity of scheduling decisions (Ettema and Timmermans, 1997). Travel is
merely seen as just one of the attributes. Moreover, decisions with respect to travel are driven by
a collection of activities that form an agenda for participation. Travel should therefore be
modelled within the context of the entire agenda, or in other words, as a component of an activity
scheduling decision. Activity-based approaches aim at predicting which activities are conducted,
where,when, for how long,with whom and the transport mode involved.
In order to summarize the most important features of activity-based modelling, we would
like to cite the work of McNally (2000), who has nicely listed 5 themes which characterize the
activity-based modelling framework:
(i) Travel is derived from the demand for activity participation;
(ii) sequences or patterns of behaviour, and not individual trips are the relevant unit of
analysis:
(iii) household and other social structures influence travel and activity behaviour;
(iv) spatial, temporal, transportation and interpersonal interdependencies constrain
activity/travel behaviour;
(v) activity-based approaches reflect the scheduling of activities in time and space.
7
3. Methodology
This section describes the methodology for the use of activity-based models in air quality
research and shows how the assessment of important variables for air-quality analysis can be
improved by the use of activity-based models. Furthermore this section presents the advantages
of an activity-based approach for policymakers who wish to reduce the impact of traffic air
pollution.
3.1 DPSIR framework
The methodology used in this research is structured according to an integrated assessment
framework known as the DPSIR framework where DPSIR stands for Driving forces, Pressure,
State, Impact and Response (figure 1). The DPSIR framework was built on a model from the
Organisation for Economic Co-operation and Development (OECD, 1993), and a few years later
adopted by the European Environmental Agency for environmental reporting purposes (EEA,
1999). This model structures the description of the environmental problems by formalising the
relationships between various sectors of human activity and the environment as causal chains of
links (figure 1).
The DPSIR approach takes into account the connections between the causes of
environmental problems, their impacts, and the societal responses to which they give rise. This
integrated assessment approach brings together relevant information about causes, impacts and
solutions in order to structure the environmental problem and present efficient policy measures of
a certain issue.
8
FIGURE 1. The Driving force-Pressure-State-Impact-Response model
The DPSIR framework can be applied to studies of the present as well as of possible
future situations. In order to structure the problem of human exposure to traffic air pollution an
adapted DPSIR framework was developed. Figure 2 illustrates the DPSIR framework for human
exposure to traffic air pollution.
DRIVING FORCES
Socio-economic and socio-
cultural forces driving
human activities
STATE
Observable changes
of the environment
IMPACT
Effects of a changed
environment
RESPONSE
The efforts of society
to solve the problem
PRESSURE
Stresses that human activities
place on the environment
9
FIGURE 2. DPSIR framework for the human exposure to traffic air pollution
Figure 2 can be explained as follows: the driving forces related to the problem of traffic
air pollution are the underlying activities that people want or have to perform. People go to work
or to school, they go shopping, have an active social life… In order to perform all these activities,
travel is a necessary consequence. Models providing predictions of the travel demand and trips
through an analysis of activities are activity-based models. The environmental pressures ensuing
from these driving forces consist of the emissions caused by the different transport modes.
Vehicle exhaust emissions contain carbon monoxide (CO), hydrocarbons (HC), nitrogen oxides
(NOx) and particulate matter (PM). The amount of pollutants depends on different vehicle and
driving characteristics, and is determined by the use of emission models. Next, the emitted
pollutants are dispersed in the surroundings leading to a degraded state of the environment.
Dispersion models aim at predicting the eventual ambient concentration of traffic air pollutants
depending on the meteorological situation and the chemical transformation of pollutants. Finally,
the exposure of people to these concentrations has an impact on human health (concentration-
response functions) and makes the society carry out a response through various actions. Through
valuation rules this approach will allow for monetization of environmental damage costs
(external costs) from traffic air pollution for human exposure (Int Panis et al, 2004).
Driving forces
Response
Indicators
Pressure
State
Impact
Pollutant emission
Ambient air pollution
concentration
Health effects
External costs
Activities
Concentration-
response functions
Environmental
modelling
Environmental
modelling
Activity-based
modelling
10
3.2 Activity-based modelling for air quality purposes
There are many transportation variables that affect emissions of traffic air pollutants. Cambridge
Systematics (1997) provides a table prioritizing the transportation data desired for emission
modelling. The following variables have been identified as the most important for this type of
analysis:
- VMT (vehicle miles of travel);
- travel by mode and occupancy rates for auto modes;
- percentage of cold/hot starts;
- speed/acceleration/driving profile;
- travel by time of day and time/location of starts;
- travel by vehicle class and model;
- travel by facility type.
The accuracy of the emissions and air-quality estimates can be no better than the
underlying transportation information (Int Panis et al, 2004). Due to the richer set of concepts
which are involved in activity-based transportation models the estimate of each of the
transportation variables listed above can be improved by using an activity-based approach
(Shiftan, 2000). Vehicle energy use and emissions depend not only on distance and the driving
speed, but also on the number of trips, the time between them, and whether the vehicle was
warmed up or not when started (Recker, 1998). The activity-based prediction of trips as parts of a
tour can identify whether a trip is a cold or a hot start. Furthermore, an activity-based model
predicts which activities are conducted, where, when, for how long, with whom and the transport
mode involved, giving accurate information about the other transportation variables in the list
like VMT, travel by mode and occupancy rate, and travel by time of day.
Activity-based transportation models not only provide accurate information on the
sources of traffic pollution but also on the location of the pollution receptors. In addition to the
advantages on emissions and air-quality estimates, the use of activity-based models also offers
opportunities to improve the exposure assessment to traffic air pollution.
11
Thus, in combination with the use of environmental models the activity-based approach is
able to integrate data on traffic activity, pollutant emission and dispersion, and travel behaviour
to derive individual- or group-level exposure measures to traffic air pollution. This integrated
method corresponds to the DPSIR approach as illustrated in figure 2. Due to the use of activity-
based models the problem related to the exposure to traffic air pollution can be split into the
different compartments of the DPSIR framework providing a thorough structured analysis of the
problem.
3.3 Air quality impacts of activity-based policies
During the past few decades, traffic has become the dominant source of air pollution in larger
cities. A lot of transportation control measures (TCM) have yet been defined to counter the rise in
vehicle emissions and energy consumption due to increased travel. Advances in technology have
played and will continue to play a role in better managing both energy consumption and
emissions associated with vehicular transport. Both environmental policy (esp. at EU-level) as
well as environmental assessments have focused on technological measures (Int Panis et al,
2003).
However, the total vehicle miles travelled and the number of trips increased drastically
lately, substantially offsetting the emission reduction through advances in technology. Other
policies, aimed at encouraging efficient trip chaining and scheduling of activities, could have the
potential to lead to significant reductions in vehicle emissions. Unfortunately the value of these
TCM strategies is often unknown as there are limited data available to measure the travel effects
of individual TCM strategies and the models are inadequate for forecasting changes in travel
behaviour that result from these strategies. The nature of the interactions among the collection of
individual and household travel decisions in response to TCM lay at the heart of the failings of
conventional models and data to provide adequate measures of their potential impact (Recker,
1998). The inability of the four-step models to evaluate responses to TCM and to provide data on
all the variables required for air-quality analysis has prompted the development of an activity-
based approach for air-quality purposes.
12
An activity-based model allows for spatio-temporal linkages between the collection of
activities that individuals and households perform as part of their daily schedule. This approach
provides a more realistic forecast of the impact of policy measures. The advantages of activity-
based modelling lie in its ability to give a better prediction of travellers’ responses to TCM and,
therefore, to provide a more accurate estimate of the changes in the important transportation
variables identified in the previous section from the implementation of TCM. One of the main
advantages of the activity-based modelling system is its ability to consider the secondary effects
of TCM. Secondary effects are adjustments to the activity pattern that have to be made in
response to the primary effect. For example, a public transport subsidy may make a commuter
change his or her mode from drive alone to public transport; this is the primary effect of the
TCM. Because, however, the person no longer drives to work, there can be no stop on the way
back to do the shopping. Therefore, upon returning home, the person takes the car and drives to a
nearby store. This is the secondary effect. In such cases, the advantages of TCM may be limited,
and the reduction of the work auto trip is offset by a new shopping auto trip. Only an activity-
based model can deal with these secondary effects.
In addition to the ability to evaluate traditional TCM such as subsidies for public transport
or for transit and adjustments to the road infrastructure, specific activity-based policy measures
also include actions like:
- changing shop opening hours;
- introducing flexible work hours;
- promoting telecommuting;
- reducing the number of cars per household.
Activity-based modelling evaluates the impact of these measures on travel behaviour and
transport, and due to this the impact on air-quality can be modelled and decisions can be made by
the policymakers. Moreover, activity-based models are able to take into account several
demographic changes to make accurate prognoses of travel behaviour in time eg. changes in
family size, number of driving licenses in the community,… providing a more realistic forecast
of the impact of policy measures in the future.
13
4. Link to sustainable mobility?
Although the list of definitions for sustainability and Sustainable Development is long, the most
widely used definition of Sustainable Development is probably the one given by the World
Commission on Environment and Developent (WCED):
Sustainable development is development that meets the needs of the present
without compromising the needs of future generations to meet their own needs
(WCED, 1987).
Sustainable Development was proclaimed a guiding political concept at the United
Nations’ World Summit on Environment and Development in 1992. The European Union has
since then adopted strategies to implement the concept in their policies. For some years now,
sustainable mobility has become an explicit aim of transport planning and policy making at local,
national and international levels. Strategies for sustainable mobility have been adopted, which
emphasise the integration of environmental and social goals with economic ones (figure 3).
FIGURE 3. The integration of environmental, social and economic goals
in the concept of sustainable mobility.
Environmental
quality
Economic
sustainability
Social
viability
14
The demand for transport services is expected to grow considerably as incomes rise, the
trend toward urbanization continues and as the process of globalisation moves forward with
expected increases in world trade and personal travel (United Nations, Economic and Social
Council, 2001). In order to meet this rising demand and because governments cannot afford to
allow transport constraints to have a negative impact on the future competitiveness of their
products, considerable future long-term investments are indispensable. In order to better guide
and substantiate the decisions of transportation planners, the use of traffic and transportation
models has been advocated by governments and by research communities.
The DPSIR approach adopted in this research reflects the multidimensional approach
inherent in the concept of sustainable mobility. It integrates data on traffic activity, pollutant
dispersion and travel behaviour, to derive individual- or group-level exposure measures to traffic
air pollution. Due to the combination of activity-based models with environmental models the
evaluation of policy measures to reduce the impact of traffic air pollution can be performed on
different levels in the framework. Even the impact of policy measures concerning the first step in
the DPSIR model, the driving forces, can be evaluated through the activity-based approach. This
methodology will allow for an accurate forecast of the impact of different kinds of policy
measures and will contribute to the improvement of a sustainable mobility.
15
5. Conclusions and further research
This paper reports on the use of activity-based models for the assessment of human exposure to
traffic air pollution. Due to the richer set of concepts which are involved in activity-based
transportation models this approach has important advantages for air quality and exposure
analyses since it integrates data on traffic activity, pollutant emission, pollutant dispersion and
travel behaviour. The use of activity-based models in combination with environmental modelling
will offer an improved evaluation of policy measures to reduce human exposure to traffic air
pollution and will allow for an accurate forecasting of the policy’s impact.
Future research consists of turning the concept of activity-based modelling for air-quality
purposes into reality. An activity-based model will be integrated with an emission and a
dispersion model to provide accurate information on the sources of traffic pollution and on the
location of the pollution receptors. First, concentrations of traffic air pollutants will be
geographically determined in a certain region. Next, the human exposure to traffic air pollution
can be assessed by combining the concentration maps with the people’s presence maps in a
geographic information system. By applying concentration-response functions and valuation
rules this approach will allow for monetization of environmental damage costs from traffic air
pollution for human exposure. An evaluation of a wide range of policy measures will be possible
and will offer important advantages for policymakers.
16
Bibliography
Ben-Akiva, M.E. and J.L. Bowman (1998). Integration of an activity-based model system and a
residential location model, Urban Studies, Vol.35 (7), pp. 1231-1253.
Cambridge Systematics (1997). Development of an Improved Framework for the Analysis of Air
Quality and Other Benefits and Costs of Transportation Control Measures, NCHRP 8-33
Research Digest.
Chapin, F.S. (1974). Human Activity Pattern in the City. Wiley, New York.
Daly A.J., H.H van Zwam, and J. van der Valk (1983). Application of disaggregate models for a
regional transport study in The Netherlands. World conference on Transport Research,
Hamburg.
EEA (1999). European Environmental Agency EEA. Environmental indicators: typology and
overview. Technical Report nº25, Copenhagen.
Ettema, K., H. Timmermans (1997). Theories and models of activity patterns, in Ettema, K. and
H. Timmermans (eds), Activity-based approaches to travel analysis, Pergamon, Oxford.
Fried, M., J. Havens, and M. Thall (1977). Travel Behavior - A Synthesized Theory, Final
Report, NCHRP, Transportation Research Board, Washington.
Hägerstrand, T. (1970). What about people in regional science? Papers of the Regional Science
Association, Vol 24, pp. 7-21.
17
Int Panis L., L. De Nocker, I. De Vlieger, R. Torfs, and S. Vermootte (2003). Diesel of Benzine?
Een onzekere vergelijking van milieuschadekosten, Arena/Het Dossier, jaargang 9, nr. 1,
pp. 6-9.
Int Panis L., L. De Nocker, E. Cornelis, and R. Torfs (2004). An uncertainty analysis of air
pollution externalities from road transport in Belgium in 2010. Science of the Total
Environment, 334-335, pp. 287-298.
Jones, P.M., M.C. Dix, M.I. Clarke, and I.G. Heggie, (1983). Understanding Travel Behavior.
Gower, Aldershot
McNally, M.G. (2000). The Activity-Based Approach. Center for Activity Systems Analysis.
Paper UCI-ITS-AS-WP-00-4.
OECD (1993). OECD core set of indicators for environmental performance reviews. OECD
Environment Monographs, n° 83, OECD, Paris.
Recker, W., A. Parimi (1998). Development of a Microscopic Activity-Based Framework for
Analyzing the Potential Impacts of Transportation Control Measures on Vehicle
Emissions. Institute of Transportation Studies. Center for Activity Systems Analysis.
University of California, Irvine.
Ruiter, E.R. and M. Ben-Akiva (1978). Disaggregate travel demand models for the San Francisco
bay area. Transportation Research Record, 673, pp. 121-128
Shiftan, S. (2000). The Advantage of Activity-based Modelling for Air-quality Purposes: Theory
vs Practice and Future Needs, In “Innovation” , Vol 13, n° 1, pp. 95-110.
Torfs, R., G. Schoeters, and P. Hoet (2004). Risico’s van inhaleerbaar stof. Tijdschrift voor
Geneeskunde, Vol 60, n°13, pp. 915-926.
18
United Nations Economic and Social Council, Commission on Sustainable Development (2001).
Transport – Report of the Secretary-General, Document E/CN.17/2001/3.
World Commission on Environment and Development (1987). Our Common Future. Oxford
University Press.