Congestion charging systems have emerged as a promising policy tool for mitigating traffic congestion and reducing emissions in urban areas. This study applies a dynamic activity scheduling model to assess the effects of congestion pricing in the Greater Helsinki region. By simulating daily activity patterns and travel behaviour, we analyse the impacts of congestion charges on mode choice, destination selection, and departure time adjustments. Our findings reveal a 10% reduction in car use and a 27% decrease in total car kilometres travelled, demonstrating the effectiveness of congestion pricing in alleviating traffic congestion. However, the analysis also highlights the potential equity concerns, with impacts varying across locations and commuting patterns. These insights contribute to the growing body of evidence on the behavioural and distributional consequences of congestion pricing, offering valuable guidance for policymakers.

Objective This study aimed to explore emergency medical dispatchers' (EMDs') experiences of prioritising patients and stewarding ambulance resources when system capacity was constrained.Design Qualitative interview study using inductive qualitative content analysis.Setting Emergency medical communication centres (EMCCs) in Sweden, operated by the national emergency call provider and responsible for receiving 112 calls and dispatching ambulances.Participants 13 purposively sampled EMDs with at least 1 year of professional experience.Data analysis Interviews were analysed inductively using qualitative content analysis (Elo and Kyng & auml;s) through open coding, grouping into subcategories and abstraction into generic categories and one main category.Results Dispatchers described prioritisation under scarcity as system work that simultaneously addressed individual patient acuity and population-level readiness. One main category captured this work: stewarding scarce response capacity. Three inter-related generic categories characterised stewardship: (1) prioritising by clinical urgency within geographic and operational constraints; (2) producing availability through anticipation, reassessment and queue governance in a 'virtual waiting room'; and (3) coordinating response through information infrastructures and interprofessional collaboration. Across categories, dispatchers described redistributing risk across patients and time, managing moral strain when delays could harm patients and using experience, reassessment and teamwork to avoid both under-response to urgent need and over-allocation that would leave areas without coverage.Conclusions Dispatch under scarcity is best understood as active stewardship of a safety-critical dispatch queue. Strengthening patient safety therefore requires organisational support for reassessment and escalation during prolonged waits, explicit governance of queue dynamics and geographic coverage trade-offs, safeguards for contextual judgement when automation is used and support for dispatchers exposed to morally difficult scarcity decisions.

In travel behaviour modelling, latent class models are used to represent underlying discrete groupings of behavioural preferences. The paper presents a latent class extension of a dynamic discrete choice model (DDCM) and applies the model to the problem of activity demand generation and scheduling. The DDCM is a recursive multinomial logit model where agents make sequential decisions in time, maximizing the expected future utility of their decisions in a random utility maximization framework. It generates activities and their associated travel within a full day schedule, endogenously respecting agents' inherent time-space constraints. The latent class DDCM builds on the base model by representing heterogeneous lifestyle preferences. A specification of the model is estimated on a Stockholm travel survey and uses age, income level, gender, car ownership and presence of children in the household as classifying variables. The models result in classes which primarily represent modality styles, finding car-, transit- and bike-primary behavioural groups as well as a multimodal group, each linked with different socio-demographic characteristics. The models improve over non-latent class reference models and provide insight into the structure of heterogeneity in travel behaviour preferences in Stockholm.

This paper presents a population synthesis model based on the Wasserstein Generative-Adversarial Network with Gradient Penalty (WGAN-GP) for training on incomplete microsamples. The proposed method aims to address the challenge of missing information in microsamples on one or more attributes due to privacy concerns or data collection constraints. By using a mask matrix to represent missing values, the study proposes a WGAN-GP training algorithm that lets the model learn from a training dataset that has some missing information. The paper compares the ability of WGAN-GP models trained on incomplete microsamples to those trained on complete microsamples to create a synthetic population. We conducted a series of evaluations of the proposed method using a Swedish national travel survey. We validate the efficacy of the proposed method by generating synthetic populations from all the models and comparing them to the actual population dataset. The results from the experiments showed that the proposed methodology successfully generates synthetic data that closely resembles a model trained with complete data as well as the actual population. The paper makes a contribution to the field by giving a strong solution for population synthesis using incomplete microsamples. It also opens up new research areas and shows how deep generative models can be used to improve population synthesis.

Background: Timely prehospital care is crucial for patients presenting with high-risk time-sensitive (HRTS) conditions. However, the interplay between response time and demographic factors in patients with breathing problems remains insufficiently understood. In this exploratory study, we applied machine learning (ML) to examine how emergency medical response time, age, and sex jointly influence the probability of encountering HRTS conditions.

Methods: A retrospective observational analysis was conducted on 132,395 prehospital missions in Stockholm (2017–2022). Multiple ML models, random forest, gradient boosting, neural networks, and logistic regression were trained to probe potential nonlinear patterns and interactions, not with the primary goal of predictive accuracy. Model performance was evaluated using sensitivity, specificity, and area under the curve (AUC) measures. However, partial dependence (PD) and individual conditional expectation (ICE) plots were the principal tools illustrating how response time, age, and sex shape HRTS likelihood.

Results: PD and ICE plots revealed that older age (> 60 years) was consistently associated with a higher probability of HRTS. Moreover, patients over 60 years displayed a complex, rising risk at prolonged response times exceeding two hours. Gradient boosting offered the best (though modest) classification metrics, with an AUC of 0.66 and an F1-score of 0.55. We emphasize that these metrics, while necessary for completeness, were secondary to our aim of characterizing nonlinear relationships.

Conclusions: Our findings underscore the exploratory value of ML in identifying subtle relationships and interactions among response time, age, and sex for time-sensitive breathing emergencies. These results highlight opportunities to refine dispatch protocols, develop age- and sex-focused screening questions, and revisit lower-priority calls after extended wait times. Future work should incorporate richer data and refine these insights for potential predictive use.

Clinical trial number: Not applicable.

Background: Emergency Medical Services (EMS) response times are critical for optimizing patient outcomes, particularly in time-sensitive emergencies. This study explores the multifaceted determinants of EMS response times, leveraging machine learning (ML) techniques to identify key factors such as urgency levels, environmental conditions, and geographic variables. The findings aim to inform strategies for enhancing resource allocation and operational efficiency in EMS systems. Methods: A retrospective analysis was conducted using over one million EMS missions from Stockholm, Sweden, between 2017 and 2022. Advanced ML techniques, including Gradient Boosting models, were applied to evaluate the influence of diverse variables such as call handling times, travel times, weather patterns, and resource availability. Feature engineering was employed to extract meaningful insights, and statistical models were used to validate the relationships between key predictors and response times. Results: The study revealed a complex interplay of factors influencing EMS response times, aligning with the study’s aim to deepen the understanding of these determinants. Key drivers of response time variability included weather conditions, call priority, and resource constraints. ML models, particularly Gradient Boosting, proved effective in quantifying these impacts and provided robust predictions of response times across scenarios. By providing a comprehensive evaluation of these influences, the results support the development of adaptive resource allocation models and evidence-based policies aimed at enhancing EMS efficiency and equity across all call priorities. Conclusions: This study underscores the potential of ML-driven insights to revolutionize EMS resource allocation strategies. By integrating real-time data on weather, call types, and workload, EMS systems can transition to adaptive deployment models, reducing response times and enhancing equity across priority levels. The research provides a blueprint for implementing predictive analytics in EMS operations, paving the way for evidence-based policies that improve emergency care efficiency and outcomes. Clinical trial number: Not applicable.

The City of Stockholm is conducting a scenario planning exercise to explore where potential future office development should be planned: closer to the city centre as in the status quo, in peripheral hubs on the outskirts of the city, or dispersed throughout multiple neighbourhoods. To support this exercise, this paper models these three scenarios using a nested work location and dynamic activity-based scheduling model. Our model predicts that high-income individuals have the highest consumer welfare benefits and are over-represented as workers in all scenarios. Developing more central office space will likely reinforce existing geographical patterns of income inequality in Stockholm; developing peripheral or dispersed office space, especially in the south of the city, will challenge these patterns. However, the model also illustrates a tension between the goals of equity and the environment. By taking advantage of existing transit infrastructure and congestion patterns, more central office development will result in lower vehicle kilometers travelled and lower car mode share for commuting than more peripheral or dispersed development.

Syftet med denna rapport är att redovisa resultat och slutsatser från en enkätstudie med intressenter inom prognosmodeller för långväga resande (resande över 100 km) som genomfördes i september 2023.

Många enkätsvar är i linje med att en framtida modell behövs för att göra ungefär samma saker som idag, att den nuvarande modellen har brister, och att bra data att skatta modellen är svåra att få tag på. Mer intressant är förslag på förändringar i modellen eller dess användning. En sådan är hur vi ska definiera en långväga resa. Det finns andra, alternativa definitioner än ”längre än tio mil” som har mer med resans ärende, start- och målpunkt eller andra egenskaper att göra.

Det finns två kommentarer, i synnerhet, med potential att påverka modellansatsen. Den första är att en prognos för ett vardags- eller årsmedeldygn inte nödvändigtvis är det mest intressanta. På vissa ställen är årsvariationen stor och dimensionerande, till exempel för turismresor, medan på andra kan dygns- och veckovariation vara viktig att förstå.

Målbaserade scenarier och backcasting nämns också. En annan typ av modell än dagens Sampers skulle kunna bidra på ett bättre sätt till att hitta lämpliga åtgärder och hjälpa till med att hitta lösningar på problem snarare än att fokusera på att räkna på effekterna av åtgärderna. Det stora arbetet är att underhålla indata och framtidsscenarier av nätverk och markanvändning. Vi skulle kunna verka för att ha en gemensam bas av indata, lämpligen också öppen för extern forskning och utveckling, som sedan kan användas till olika modeller beroende på vilken fråga som ställs.

När det gäller kompetens tolkar vi intervjusvaren som att den behöver stärkas både bland användare så att resultaten kan genereras och tolkas med förståelse av orsakssamband, och bland modellutvecklare. Samtidigt upplever en del respondenter att systemet är tungarbetat och det används i praktiken mest för ett relativt fåtal scenarioanalyser. Ibland skulle det nog vara mer användbart med ett prognosverktyg att testa idéer eller alternativa åtgärder i ett tidigt skede. Det finns också svar som antyder en efterfrågan på en modell som kan bidra tydligare till problemlösning och målstyrning. Dagens mer prognosinriktade modellansats kan användas till sådant, men inte på ett naturligt sätt.

Vår slutsats av intervjustudien är att den har gett oss insikter om användarnas uppfattningar och behov. För att utveckla framtidens prognosmetodik behöver vi komplettera med kunskap om nya metoder och dess egenskaper i en litteraturstudie.

The aim of this report is to report results and conclusions from a survey study with stakeholders within forecast models for long-distance travel (travel over 100 km) that was carried out in September 2023.

Many survey responses are in line with the fact that a future model is needed to do roughly the same things as today, that the current model has flaws, and that good data to estimate the model is hard to come by. More interesting are suggestions for changes to the model or its use. One such is how we should define a long-distance trip. There are other, alternative definitions than "further than one hundred kilometres" that have more to do with the purpose of the journey, the origin and destination point or other characteristics.

There are two comments, in particular, with the potential to influence the model approach. The first is that a forecast for a weekday or annual average day is not necessarily the most interesting. In some places, the annual variation is large and dimensioning, for example for tourism trips, while in others understanding the daily and weekly variation can be important.

Target-based scenarios and back casting are also mentioned. A different type of model than today's Sampers could contribute in a better way to finding appropriate measures and helping to find solutions to problems rather than focusing on calculating the effects of the measures. The major work is to maintain input data and future scenarios of network and land use. We could work to have a common base of input data, preferably also open to external research and development, which can then be used for different models depending on the question being asked.

In terms of competence, we interpret the interview responses as needing to be strengthened both among users so that the results can be generated and interpreted with an understanding of causal relationships, and among model developers. At the same time, some respondents feel that the system is heavy-duty and that it is used in practice mostly for a relatively few scenario analyses. Sometimes it would probably be more useful to use a forecasting tool to test ideas or alternative actions at an early stage. There are also answers that suggest a demand for a model that can contribute more clearly to problem solving and goal management. Today's more forecast-oriented model approach can be used for such, but not in a natural way.

Our conclusion from the interview study is that it has given us insights into users' perceptions and needs. To develop future forecasting methodology, we need to supplement with knowledge of new methods and their characteristics in a literature study.

Cost-benefit analysis (CBA) for cycling infrastructure investments are less sophistically developed compared to the ones for private cars and public transport, and one of main reasons is the lack of “well-developed” transport models for cycling. In this study, a dedicated transport model for cycling is used to appraise cycling infrastructure investments in Stockholm, Sweden. The model captures the impact of a change in cycling infrastructure on cycling route choice, mode choice, destination choice and trip generation and calculates cycling flow on link level. the generalised cost measure defined in the route choice model captures the impact of cycling infrastructure. Results suggest that although cycling flow on the links with investment may increase substantially, only a small share comes from modal shift and thus the external effects such as reducing car congestion and emissions are marginal. For all three scenarios investigated, over 97% of the benefits measured in the unit of generalised cost belong to the existing cyclists. The route choice model does not minimize travel time but generalised cost which also measures health, safety benefits and other possible benefits that may be considered by the cyclists when they choose to cycle. In fact, travel time saving benefits of the investments evaluated in this paper are all negative. The existing effect evaluation models therefore need to be adjusted to be more consistent with the transport model.