Sufficiency in passenger transport and its potential for lowering energy demand

Transport systems around the world are moving people and goods faster than ever before, reaching new performance peaks every year (Mattioli and Adeel 2021, OECD 2023). This growth poses several sustainability threads: Particulate matter pollution from traffic causes autoimmune diseases and generates high costs in healthcare systems (Levy et al 2010, BAFU 2020), traffic accidents account for the greatest proportion of deaths among young people (Peden et al 2022), and transport infrastructure destroys local ecosystems in rural regions (Sovacool et al 2021) or is socially inequitable in urban regions (Creutzig et al 2020). The Paris Agreement's climate change mitigation targets further put the shift towards sustainable mobility under serious time pressure. North America's and Europe's transport activities alone consume 30% of global final transport energy demand (IEA 2022), which is disproportional to their population share. Thereof, passenger transport consumes 68% (European Commission 2021, BTS 2022). Today, this causes over-proportionally high greenhouse gas (GHG) emissions. By extrapolating this trend, most long-term decarbonisation studies still assume increasing transport activity levels and energy demands for high-income countries, even those considering the 1.5^∘ target (C1 and C2 scenarios in the IPCC AR6 (Byers et al 2022)). However, recent studies show that lowering final energy demand is crucial for achieving this goal (Grubler et al 2018, Keyßer and Lenzen 2021). Still, this field stays under-researched, especially for high-income countries, where future demand assumptions spread wide between unconstrained growth and radical decline.

A common term addressing these challenges in passenger transport is sustainable mobility. There are different definitions from various research fields, which allow for rather flexible interpretation of this concept (Berger et al 2014, Gallo and Marinelli 2020). Generally, and especially in the context of climate change mitigation, the vague goal of sustainable mobility commonly translates into three more tangible strategies (see Banister 2008, Nykvist and Whitmarsh 2008, Berger et al 2014): Avoid the need for traffic, Shift traffic to more sustainable modes, and Improve transport technologies. In long-term decarbonisation studies, technologically dominated Improve measures, such as high market penetration of electric vehicles, have already seen a lot of attention, while Avoid and Shift measures remain underrepresented (Gota et al 2019). However, the rigid time limitation for limiting global warming to 1.5° requires unleashing full transport decarbonisation potential (Mundaca et al 2019).

Hence, we focus on Avoid and Shift measures and frame them as demand-side mitigation strategies, contributing to the ongoing debate in scientific literature (Sorrell et al 2020, Creutzig et al 2022). We exclude Improve measures that require behavioural adoption towards new technologies by end users, but do not affect the service provided (e.g. electric vehicles provide the same service as combustion engine vehicles, even though they differ in energy service). On this basis, we draw on definitions of energy sufficiency (Zell-Ziegler et al 2021) and sustainable consumption corridors, framing Avoid and Shift measures as sufficiency measures (as derived in appendix A). We focus on passenger transport not only because of its high energy demand, but also due to its complex cultural embedding (Mattioli 2016), which has not been analysed thoroughly in the context of sufficiency transitions. Freight transport would require a different approach, as it is a secondary effect of consumption patterns, industry facilities, and transportation cost.

How can sufficiency look like in passenger transport? Which influence do sufficiency futures have on the transport system? Some modelling exercises tried to answer parts of these questions (e.g. Anable et al 2012, Replogle and Fulton 2014, Pomponi et al 2021), but there is no reference of sufficiency in mobility describing a comprehensive transition process to the best of our knowledge. We choose an interdisciplinary, participatory research design to analyse a sufficiency-directed transport system design in Germany (section 2). We consider the spatial restriction important for a detailed examination of sufficiency transitions, as they are context-specific (Sandberg 2021). Germany is particularly interesting as a high-income country with strong car-dependency and growing transport activity while population is declining (BMDV 2021).

Answering these research questions about sufficiency futures requires a normative approach, as it implies countering currently observable trends (Banister and Hickman 2013). We employ a backcasting scenario technique, where we start from a desirable end-point and examine the means by which this future can be attained (Robinson 1982). Transport modellers commonly use this technique to analyse policy pathways. However, sufficiency transitions—the process of advancing sufficiency practices—are multidimensional processes (Sandberg 2021), which require methods of analysis that go beyond common policy analysis and (techno-)economic modelling practices. Schwanen et al (2011) argue that the qualitative basis which ensures valid quantitative outcomes should receive more attention in transport research. Additionally, a participatory perspective is deemed important to explore energy demand perspectives in its full complexity (Hirt et al 2020, Nikas et al 2020).

Some studies addressed these points for passenger transport decarbonisation in the past: Köhler et al (2020) combine qualitative mobility narratives, derived with the multi-level perspective (MLP) (Grin et al 2010), and an agent-based model in order to describe the Dutch low-carbon mobility transformation. A Danish study of transport and energy system decarbonisation uses a participatory and narrative-based research design, following the Story and Simulation approach (Alcamo 2008), in order to model long-time policy scenarios (Venturini et al 2019). Anable et al (2012) constructed a lifestyle storyline and quantified its transport demand in a spreadsheet model. Several other approaches exist in the domain of transport modelling (e.g. Hickman et al 2012, Banister and Hickman 2013, Varho and Tapio 2013), where participation of stakeholders is very common, but use of transition theory is rare. Building upon this work, we use methods from socio-technical transition theory together with aggregated transport modelling, wrapped into a participatory research design. It divides in two main phases: development of storylines as a qualitative basis and scenario modelling yielding the quantification of storylines (see figure 1). We do not consider a temporal dimension, as our quantitative method simulates only one year with the desired transport system configuration and our qualitative method does not allow for exact identification of a temporal scope. In the end, the time frame depends on the level of ambition, but is restricted by realism.

Zoom In Zoom Out Reset image size

2.1. Storyline development

2.2. Translation and quantification

2.3. Transport modelling

2.4. Reflection on methods

The Avoid and Avoid+Shift scenarios show the strongest total decline in commuting and business trips due to remote work and remote meetings throughout industries (figure 2). Non-compulsory trip purposes are less prone to decrease, but still decline due to increasing digitisation of social events and shopping/execution trips. Here, particularly trips in rural and suburban areas, that go beyond the municipality borders (i.e. inter-zonal trips), decline because of spatial planning processes that increase the diversity and livelihood in the local environment. Local economies, moreover, boost the availability of goods, services, and amenities. The decline in inter-zonal trips is especially interesting, as they produce the gross of total passenger kilometre (pkm; 86%), although being responsible for only 36% of total trips in the reference scenario. The Shift scenario exhibits a slight increase in non-compulsory trip purposes, which are sensitive to lower travel cost (time and price). Compulsory trips, however, do not see an absolute increase, but a shift from car-owning households to non-car owners: Car dependency declines to a minimum and everybody gets full access to the labour market and education system ⁵ (likewise in the Avoid+Shift scenario).

Zoom In Zoom Out Reset image size

Car ownership (defined as households with at least one car available) and underlying drivers vary largely between scenarios (see table 2). The Shift scenario features a highly interconnected PT system, but reduces car ownership only by 2% through more attractive transport services and ticket offers for older generations. Hence, the social status of cars prevails, even though its use decreases. In contrast, the Avoid scenario reduces mileage and car ownership through a trend 'from ownership to access'. It features widespread opportunities for car sharing on the one hand and increased local cohesion on the other hand (e.g. through mixed neighbourhoods, revival of hitchhiking, and local economies), reducing the need for private cars. Adding stringent tax policy and multiple other car-disincentivising policies, the Avoid+Shift scenario exhibits an average car availability of 52% (only 29% in cities). It entirely shifts the perception of cars towards an anti-status symbol, drastically reducing use and ownership.

Scenario trends in car ownership reflect themselves in modal shares, too. Figure 3 shows that modal shares of rail and bus transport can increase by 30% when PT supply drivers of the Shift scenario interact to make it as attractive as possible. There is also a notable increase in cycling, mainly driven by significant investments into cycling infrastructure and safety, as well as cultural shifts towards tranquility and appreciation of the landscape. Especially car owners use the bicycle more for leisure and execution trips. Vice versa, car sharing—as introduced in the Avoid scenario nationwide—is used only by non-car owners, as the private car is still cheaper in operation. Car sharing helps reducing car ownership, providing the feeling of flexibility, even though it is used only occasionally.

Zoom In Zoom Out Reset image size

The Avoid+Shift scenario reduces car transport by 46% in total, compared to the reference. This is due to drivers pulling users towards PT (as in the Shift scenario), drivers pushing users away from cars (e.g. car-free inner cities or tax policy), and drivers that affect mobility culture (e.g. ban of car advertisement or trends of post-materialism). Still, the private car persists as an important pillar of the transport system and is used for all trip purposes, especially on short distances. Where Shift drivers apply, medium to long distances are dominated by PT, as it offers overly attractive connections and tariffs (see figure 3). The degree of urbanisation reduces the importance of the car, but even a fraction of city inhabitants keeps using it for daily short-distance trips. In fact, the average distance slightly decreases in the Avoid and Avoid+Shift scenarios for private cars, while car sharing's average distance pertains at the reference level, as shared cars are used for holiday trips, too.

Figure 3 also depicts the trip distance distribution. The Avoid and Avoid+Shift scenarios decrease total trip frequency and shift short-distance trips to the local environment, where availability of daily demands increases. On the other hand, the Shift scenario shifts medium-distance traffic to shorter distance classes. This is due to local on-demand services which replace scheduled bus services outside of cities and make short-distance travel overly attractive: They reduce the average access, egress and waiting time significantly and increase PT flexibility by offering efficient ride-pooling schemes. They dominate road PT with a 90% share of trips.

The perceived travel cost, which consists of transport system prices, travel times, and mobility attitudes of the corresponding demand segment, is an economic measure from transport research. It shapes the choice of trip destinations, i.e. the distance distribution, as can be seen in figure 3. The Shift scenario exhibits a smoother ascent than the reference due to lower PT fares and better connectivity. In contrast, the Avoid+Shift scenario imposes high cost on car travel, yielding a steeper curve. Here, the curve is smoother for the rural population and steeper for urban inhabitants, as rural environments become more accessible and cars are pushed out of cities, respectively. In general, scenarios with Avoid character shift attitudes from appreciating inter-zonal travel towards local cohesion.

Resulting passenger kilometres are similar to mode shares (see figure 4). In general, urban-rural mobility differences decrease in the sufficiency scenarios. Still, urban regions are better connected by long-distance train services and cycling highways in the Shift scenarios, leading to less car travel. Compared to the final energy demand, road and air transport stand out, as they are less energy efficient than rail modes. These figures imply a technology mix that is 100% electrified, as described in section 2.2. Hence, carbon emissions depend on the carbon intensity of the electricity mix and go beyond the scope of this study.

Zoom In Zoom Out Reset image size

The three sufficiency scenarios are the product of qualitative storylines and the quantification of their drivers of change. All corresponding modelling results support key features of the storylines (see appendix B and table 1). The Avoid scenario makes 38% of motorised trips obsolete through sufficiency-directed spatial planning, digitisation of work relations, and cultural, as well as economic trends forming a 'new localism'. About 6% more trips can be avoided in the Avoid+Shift scenario by dis-incentivising car travel; especially on long distances through high taxation and in cities through regulation. The new perception of cars as anti-status symbol and corresponding policy measures also reduce their mode share by 13% more than the Shift scenario can achieve. The latter represents a classical pull strategy ⁶ with a highly attractive PT system and growing PT culture, but no shift in economic principles.

It is questionable, whether the Shift scenario is a sufficiency scenario after all. There is no consistent definition of sufficiency in mobility, but in general, sufficiency transitions are a fundamental change towards 'enoughness' (Jungell-Michelsson and Heikkurinen 2022), while some connect it to degrowth (Lage 2022). The Shift scenario shows no such change, as for individual consumption, and even produces more trips than the reference. It follows liberal principles of increasing transport connectivity that used to dominate transport planning of the last century (Marvin and Guy 1999). Still, the Shift scenario fits our notion of transport sufficiency from an energy demand perspective, as it shows a 30% reduction while using the same propulsion technologies. Further research on sufficiency should gain a deeper understanding of fundamental systemic change that facilitates sufficiency transitions to define this concept more precisely.

Nevertheless, the Avoid+Shift scenario satisfies all sufficiency requirements. On the one hand, it reduces mobility consumption, on the other hand, it dis-incentivises car travel and stimulates PT use—pull and push. It establishes a new mobility culture by setting new fundamental principles of mobility: equity, health, and diversity (see appendix B). This is highly effective, as can be seen in mode shares and trip frequencies. These indicators are malleable through diverse political, economical, technological, and social interventions, as well as their interplay (see table 1). More precisely, we observe synergies. Single policy measures like road charges show a lower impact on the modal split as when combined with PT investments. Vice versa, infrastructural measures should be accompanied by regulatory interventions to yield the maximum outcome: Mobility hubs at city borders show no effect, unless accompanied by car-free inner cities; cycling highways are not as attractive without the ban of car advertisement. All developments in these comprehensive sufficiency scenarios show higher effects than the sum of its single measures, as decision processes are non-linear. Hence, the combination of pull and push measures, as well as political, economical, technological, and social drivers is crucial for promoting sufficiency. There is no 'silver bullet' in transport policy (see Givoni et al 2013).

Still, our scenarios show two measures with especially high impact. An important pull measure is full coverage of remote regions with on-demand ride pooling services. Such services have seen much attention in recent research (Maas 2022). If they are managed highly convenient, they will help fostering a public transport culture. Rail transport can profit from this culture, but has to become reliable and operated at higher frequency. Moreover, high-density living blocks in agglomeration areas show large reduction potentials in motorised transport—especially private car use –, if they replace (newly built) single-family home settlements to counter current trends. High-density living comes with lower resource use, which supports sustainability ambitions in the building sector.

Another important discussion regarding sufficiency transitions is quality of human life. Here, we draw on eudaimonic notions of well-being, as it is founded on universal, objective, and structural conditions that facilitate a good life (Lamb and Steinberger 2017). Rao and Min (2018) utilise this concept to quantify decent material living standards, which Millward-Hopkins et al (2020) use to quantify decent energy needs. Its comparison in figure 4 shows that all German sufficiency scenarios are well above the minimum energy demand necessary to provide decent levels of mobility. But well-being also accounts for other indicators that see improvements in our scenarios: Public health, social equity, accessibility to need satisfiers, and social cohesion in the local environment. This is consistent with global-level findings from Creutzig et al (2022), who underline synergies between sufficiency measures and well-being, and Roy et al (2021), who find positive effects of sustainable mobility on several Sustainable Development Goals. Further research should define and quantify impacts on well-being indicators in mobility (Zhao et al 2020), which are, to this date, not clearly defined (Virág et al 2022). However, we show that well-being at high levels of personal mobility does not depend on growth in pkm.

The final aim of this study is to explore the energy demand reduction potential of sufficiency futures, which makes the realisation of ambitious climate targets more likely. All scenarios reduce energy demand—up to 73% in the Avoid+Shift scenario. These 2.04 GJ of final energy per capita in 2050 are also 73% lower as the Global North's land transport final energy demand per capita in the low energy demand (LED) scenario (Grubler et al 2018), which is the IPCC Sixth Assessment Report's most ambitious Illustrative Mitigation Pathway. It shows 29% increase of transport activity in global north countries (20% reduction against historic trends) and moderate shifts from car to rail and air transport. Mode shifts are mainly driven by developments towards teleworking and more liveable and healthier cities. The LED scenario halves the vehicle stock through car sharing and digitisation of mobility supply, which is a relevant lever for transport's total energy demand.

Our Avoid+Shift scenario reduces the private vehicle fleet by 36%, although the low average mileage suggests that car ownership could drop even further in the long run. Put together, the LED scenario's narrative differs from our sufficiency storylines. We do not have the same focus on digitalisation, even though it is a major driver for traffic avoidance in work relations and a crucial enabler of new technologies that resolve car dependency and corresponding inequity. The LED narrative breaks with fewer conventions than the Avoid+Shift storyline, yet it is difficult to assess which storyline is more likely. The transition dynamics and driver classes in table 1 suggest that our storylines rely to relatively large degrees on target-oriented policy intervention. Individual mindset shifts are largely enabling these actions, but also resulting from them. Corporate action is mostly a result of the new system configuration. Consumption changes towards local products and services are an enabler of local economies, even though economic system shifts require more profound drivers.

Results of this paper contrast common transport demand-side assumptions in long-term energy modelling (see Byers et al 2022). They suggest that more ambition in reducing energy demand through demand-side interventions is possible and show a promising future for both, climate change mitigation and human well-being. These results can incentivise global energy modelling studies to consider behavioural change more thoroughly, as it is often overlooked today (Samadi et al 2017). That might go beyond current economic principles and challenge the economic growth dependency, a topic that should receive more attention in the IPCC Seventh Assessment Report (Keyßer and Lenzen 2021). However, this is just a national, sectoral study. More research should direct towards its connection with the energy system, the freight transport system, and towards exploring comprehensive sufficiency transitions for the entire economy, and in other regions of the world.

We sincerely acknowledge funding from the Reiner Lemoine Foundation within the 'EnergieSystemWende' Graduate School, as well as the support given by the German Federal Ministry of Education and Research (BMBF), which funded the 'Coal-Exit' research group [Grant Number 01LN1704A] under the 'Global Change' funding priority. We also acknowledge support by the German Research Foundation and the Open Access Publication Fund of TU Berlin. Moreover, we are very thankful for dedicated support by D Süsser, P Blechinger, and two anonymous reviewers, who helped advancing this study substantially, as well as support in model refinement by J Thema and J Heimrich.

The transport model used for this study is openly available on github: Arnz (2023). The repository contains all data relevant for this study. The supplementary material contains further information used to create the sufficiency storylines and scenarios.

In 2022, the Working Group III 'Mitigation of Climate Change' of the IPCC Sixth Assessment Report for the first time dedicated special attention to demand-side mitigation measures, i.e. climate change mitigation measures aiming at changes in energy and material consumption patterns. This is due to a growing body of literature concerning demand-focused interventions and sufficiency (as a complement to the sustainability categories efficiency and consistency). Despite this evolving research, the definition of sufficiency is context-dependent and remains unclear when looking into a sub-system such as passenger transport.

Generally in scientific literature, sufficiency concerns the question of 'how much is enough?'. Spengler (2016) connects two kinds of 'enoughness' as lower and upper boundary conditions of sustainable human life and thereby connects sufficiency to the common concept of sustainable consumption corridors (see Raworth 2017): The lower limit refers to human needs and a minimum approach to distributive justice, while the upper limit concerns not exceeding planetary boundaries. The political challenge is governing in between these boundary conditions (Spangenberg 2014) to foster a good life for all (O'Neill et al 2018).

In transport literature, there is no uniform definition of sufficiency, even though there are overlaps with Avoid and Shift strategies of sustainable mobility. A growing body of literature treats sufficiency as an individual attitude to mobility behaviour that helps reducing GHG emissions (e.g. Verfuerth et al 2019, Vita et al 2019, Loy et al 2021). Zell-Ziegler et al (2021) draw on the concept of energy sufficiency and define sufficiency in transport as a change in service quality yielding lower energy demand, facilitated mainly through Avoid and Shift measures. Waygood et al (2019) define transport sufficiency from an urban planning perspective as 'to achieve the best quality of life given global constraints'. Here, quality of life does not only refer to benefits of carrying out human activities, which require physical mobility. It also refers to negative impacts of transportation and large potential co-benefits of their mitigation (see Creutzig et al 2022).

We build upon these notions of sufficiency from a systemic energy demand perspective. We treat transport sufficiency as a system design strategy to support a swift reduction of energy demand and GHG emissions. This concerns the upper boundary condition of sufficiency. The lower boundary, basic human needs, is more fuzzy because mobility serves as an indirect need-satisfier. We define basic needs as independent of the mobility culture, which is relative, contextual and historical (Mattioli 2016). Hence, the immediate action, like car driving, is no need on its own. Virág et al (2022) try to define 'decent mobility standards', which describes decent levels of physical mobility connected to decent levels of well-being across different cultural and spatial contexts. Even though they cannot find a definite threshold, we adopt this idea as lower boundary of transport sufficiency. We can compare our results to the absolute lower bound of sufficiency as calculated in Millward-Hopkins et al (2020), who are not taking into account the cultural context and the current built environment.

The Shift storyline describes a radical pull strategy that reduces car dependency to the absolute minimum. Drivers of this process mainly concern strengthening of PT and cycling, transport planning, and digitised mobility services. Rapidly increasing the reliability and capacity of the rail network is of particularly high priority, as well as the establishment of comprehensive on-demand ride-pooling systems. The necessary money comes from the expansion stop of roads and airports, comprehensive parking pricing in public urban and suburban spaces, and the reduction of climate-damaging or car-friendly subsidies. Transport planning and the corresponding budget is fully directed towards PT. As juridical underpinning, road traffic regulations give priority in the traffic flow to cycling and PT. Moreover, wide and secure cycling highways between urban and suburban regions incite more active mobility.

Technological and organisational innovations, such as mobility hubs in metropolitan areas, free bicycle entrainment, or bike sharing hubs at train stations, enable comfortable multi-modality. E-bikes and cargo bikes support the shift to active mobility additionally. Digital mobility services are emerging that are easy to understand, include all transport services and can be used nationwide. In any case, all PT schedules are well coordinated and a uniform tariff system throughout Germany facilitates easy use and reduces prices in remote regions. There, too, and at off-peak times, autonomous on-demand shuttles provide high service quality and flexibility.

This completely new prioritisation also involves a lot of education work among the population. The most important actor here is the federal government in cooperation with local transport planning. Besides these strong top-down initiatives, innovative business models are also driving the process. Society plays a minor role and adjusts its mobility culture to the new transport system with a temporal delay.

The Avoid storyline describes cultural and economic change resulting in traffic avoidance—eliminating the need for long and many trips. The initiative comes from two different directions: Top-down and bottom-up. Urban and spatial planning focuses exclusively on densification of existing settlement areas instead of new development, as well as improvements in quality of life and diversity in the local habitat. In this way, many journeys by motorised vehicles become unnecessary, because the environment in walking distance offers shopping, errands and recreational opportunities, as well as space for social activities.

At the same time, various new bottom-up initiatives establish local economies and restorative, local lifestyles. The former strengthen local coherence and make the decentralised offer of products, services, and amenities economically attractive. The prerequisite for this is a less growth-oriented economic policy and rejection of materialism throughout a critical mass in society. Car sharing systems, which emerge throughout the country, support this trend: Following the slogan 'from ownership to access', they lead to reduced car dependence and ownership. The new lifestyles are characterised by local cohesion, while social contacts and work relationships that lie outside the local area are primarily cultivated in digital space. To this end, structures and rights for remote work are comprehensively created and their conditions favoured. This goes beyond office work and comprises remote control of industrial sites.

The Avoid+Shift storyline combines three elements: the radical pull strategy of a reliable and interconnected public transport system (as in the Shift storyline), resilient local lifestyles (as in the Avoid storyline), and a fundamental restructuring of transport planning and economic activity. While the first storyline minimised car dependency, the fundamental restructuring aims at maximum human-centred mobility planning and minimum car ownership (as a main driver of transport externalities), facilitated through strong top-down initiatives and large-scale shifts in individual mindsets.

The regulatory framework is subject to particularly strong adaptation. Interdepartmental mobility policy bans cars from cities, leads industrial policy to the necessary shift towards PT, bans car advertising (because of the severe consequences for health and life) and reforms the tax system to incentivise PT use and disincentivise car ownership. Hence, ousting of the automobile lobby from party politics is necessary. At the same time, transport planning becomes more people-centred, more diverse, better staffed and better integrated with urban and spatial planning. Its guiding principles are equity (in terms of reducing car dependency), health (fostering active mobility and lowering transport externalities), and diversity (including all perspectives of society into transport planning).

On a cultural level, climate protection, social justice and health—corresponding to the new mobility planning principles—are becoming more important in the consciousness of the population, while economic growth and materialism are losing relevance. Comprehensive criticism of consumption manifests itself in sufficiency-oriented lifestyles, which is demonstrated by role models from the rich and influential classes and slowly spreads through all layers of society. Socially, the car is not only losing its status, but is becoming an anti-status symbol for a critical mass of the population. This is made possible by an ongoing global restructuring of the economic system with the aim of decoupling prosperity from growth, as the neo-liberal economic system is coming under strong pressure due to the consequences of climate change.

We refined the demand model structure, as initially described in Arnz (2022), in order to endogenously depict generation and distribution of trips for each demand segment. Compulsory trips (i.e. commuting, education, and business trips) are computed using a doubly constrained distribution with the logsum of mode choice utility building the deterrence matrix. Trips for other purposes (utilities, leisure, and accompany) utilise multinomial logit models to depict trip generation and destination choice, respectively. The generation model's utility function looks as follows:

Applied to zones z for each demand segment i: with and without car availability for each non-compulsory trip purpose. Choice alternatives describe the number of trips per day. Except for j = 0, all alternative-specific constants are fixed to zero. Zone population , average household size , household income , and the population share of a certain occupation (; not for buy/execute trips) influence the decision. Moreover, the trip frequency depends on the accessibility acc (calculated as the average cost of mobility to other zones), linking the generation of trips to the transport system design. Marginal utility parameters α to η for every alternative and demand segment are calibrated using the same mobility survey as the mode choice model ('Mobility in Germany') - here and in all following choice models. Building upon this, a binary logit model formulates the choice between executing a trip within or beyond the origin zone's boundaries:

with

While inter-zonal choice utility depends on the zone's accessibility and an , inner-zonal utility consists of the zone's population density and the number of attractions a_n of the attraction categories Aⁱ that are relevant to this demand segment. Corresponding points of interest data for these categories was fetched from OpenStreetMap in 2022. For the choice between inter-zonal trip destinations, a third-level logit choice model is applied. Its utility function concerns the same demand segments and zones, while choice alternatives d comprise the full set of model zones:

with , following the formulation for destination choice models with attraction variables from Daly (1982). Additionally, the distance between origin and destination and the squared distance are significant choice variables. Here too, cost of mobility CC (i.e. the mode choice composite cost) influence the distance distribution of trips. Resulting trip volumes are calibrated towards distance distributions from the national German mobility survey and total volumes of the Federal Ministry of Transport (BMDV 2021). Additionally, we implement interfaces for drivers that affect model parameters, as described above. The full list of drivers and brief descriptions of their implementation can be found in the supplementary material.