Road traffic noise is a major environmental pollutant with significant societal impact, warranting reliable exposure assessment. Beyond its health impact, noise also contains rich temporal features that can be exploited for analysis. The state-of-the-art CNOSSOS-EU framework requires detailed traffic data, but in cities with limited monitoring infrastructure this input is often unreliable, reducing assessment accuracy. The thesis addresses this gap by developing a methodology that exploits data from roadside noise sensors and expands their functionality to also serve as traffic sensors, providing the necessary inputs to a modelling chain for road traffic noise.

The methodology is validated through a case study on a busy urban road in Stockholm, Sweden. Three primary objectives are considered: (i) estimating traffic flow rates from noise measurements, (ii) evaluating the use of noise data as an alternative to radar in a microscopic noise assessment framework, and (iii) developing a methodology to define noise mitigation strategies using city-wide noise assessments.

Ad-hoc noise sensors were deployed at three positions at the case study location for over 400 days, generating data for the development and testing of traffic flow estimation models. These models were also adapted through specialized training to reduce the need for expensive traffic sensor data when estimating conditions at a new location. A microscopic-traffic-based noise simulation framework was implemented and run with input from noise data, and its outputs were compared against those from radar-based input. Algorithms for optimal vehicle routing were combined with noise-based cost functions to determine vehicle routes reducing population exposure to noise.

Results show reliable traffic flow estimation across temporal and spatial variations. Combining datasets from multiple locations and using synthesized data show potential for cost-effective implementation at new locations. Output of noise simulations based on noise measurements show good agreement with output based on radar data. Inclusion of noise exposure constraints in vehicle routing identifies routes with lower noise exposure, while remaining logistically feasible.

The findings demonstrate the potential for noise sensors to serve as a reliable traffic sensor, and drive noise simulations with accuracy similar to that from radar data. These low-cost multifunctional sensors are well suited for large-scale deployment in smart cities, supporting noise assessment and mitigation, as well as systems for intelligent traffic management.

Vägtrafikbuller är en betydande miljöfaktor med stor samhällspåverkan, vilket motiverar en tillförlitlig exponeringsbedömning. Det nuvarande CNOSSOS-EU-ramverket kräver detaljerade trafikdata, men i städer med begränsad övervakningsinfrastruktur är sådan information otillförlitlig, vilket minskar bedömningens noggrannhet. Denna avhandling tar sig an denna brist genom att utveckla en metod som utnyttjar data från ljudnivåmätare utplacerade längs vägar, och utökar användingsområdet så att de även fungerar som trafiksensorer, vilket ger nödvändiga data för modellering av trafikbuller.

Metoden valideras genom en fallstudie på en trafikerad gata i Stockholm. Tre huvudsakliga mål beaktas: (i) uppskattning av trafikflöden utifrån ljudmätningar, (ii) att jämföra användningen av ljuddata som ett alternativ till radar för bullerbedöming på mikronivå, och (iii) att utveckla strategier för dämpning av buller baserat på bedömningar för hela staden.

Specialkonstruerade ljudnivåmätare placerades på tre positioner på fallstudieplatsen under mer än 400 dagar, vilka genererade ljuddata för utveckling och testning av modeller för trafikflödet. Modellerna tränades sedan och anpassades för att minska behovet av data från resurskrävande trafiksensorer vid uppskattning av trafikflödet på en ny plats. Ett ramverk för simulering av trafikbaserat buller på mikronivå implementerades och testades med input från uppmätta ljudnivåer, och resultatet jämfördes med radarbaserade resultat. Algoritmer för optimering av fordonsrutter kombinerades med ljudbaserade kostnadsfunktioner för att fastställa rutter som minskar befolkningens bullerexponering.

Resultaten visar på möjligheten för tillförlitlig trafikflödesberäkning som tar trafikens tidsmässiga och rumsliga variationer i beaktande. Kombinationen av mätdata från flera platser och användningen av syntetiserade data visar på potentialen för kostnadseffektiv implementering för nya platser. Resultaten från bullerberäkningar baserade på ljudmätningar visar god överensstämmelse med resultaten baserade på radardata. Begränsningar av bullerexponering i fordonsruttplanering visas kunna ge rutter med lägre exponering, samtidigt som de förblir logistiskt genomförbara.

Resultaten visar att ljudnivåmätare kan vara tillförlitliga trafiksensorer och ge underlag för ljudsimuleringar med en noggrannhet som är likvärdig med den som uppnås med  radardata. Dessa kostnadseffektiva multifunktionella sensorer är väl lämpade för storskalig användning i smarta städer, som ett verktyg för bedömning av och åtgärder för minskat buller och stödja ett system för intelligent trafikledning. 

This study proposes the determination of a log-linear regression model for estimating average traffic flow rates using a single measured noise indicator. This model was trained and tested with noise and traffic count data collected over 400 days at a case study location in central Stockholm, Sweden. Through a comprehensive analysis of the correlation between various noise indicators and traffic counts, the best performing indicator was selected, enabling traffic flow estimation with an average day-wise RMSE of 2.31 vehicles per minute and percentage error of 7%. Different measurement campaign strategies were tested to assess their effectiveness in providing reliable training data, demonstrating that campaigns measuring over all hours of the day and all days of the week perform significantly better than campaigns restricted to typical weekday working hours. This study highlights the potential of noise-based traffic estimation as a complementary, cost-effective approach for enhancing real-time traffic monitoring and transportation assessment.

Road-traffic flow parameters are a critical input to urban noise pollution evaluation frameworks such as CNOSSOS-EU. The accuracy and resolution of the traffic flow improves the quality of the noise exposure assessment. Highly resolved traffic flow within an urban network is difficult to reliably measure or predict. The quality and availability of such traffic information may be improved using data from low-cost noise sensors that are mounted beside representative segments of the road network. A model is presented, which relates the noise levels measured at a single noise sensor with the local traffic flow conditions. The noise data is obtained from a noise sensor mounted alongside a road, which is part of a test-bed in Stockholm, Sweden. The local traffic flow is obtained from commercial vehicle-counting sensors mounted alongside the noise sensor. Effect of seasonality is observed when comparing models trained on different subsets of the training data. The effect of seasonality on the noise levels is investigated.

Road traffic noise calculations require modeling the traffic flow in a road network. The reliability of these calculations can be improved with accurate estimation of the traffic flow, including estimation of its temporal variations. Low-cost noise sensors that run on single-board computers in a noise monitoring network are suitable candidates to simultaneously estimate the local temporal traffic flow from their pass-by measurements, using an on-board traffic flow estimator model. Aside from this model requiring to be computationally efficient, it should also be robust, e.g., invariant to sensor position relative to the source, weather conditions, etc. With noise measurements as an input, different noise features and prediction models are tested for vehicle detection. The accuracy of these models is evaluated using traffic count data obtained from dedicated vehicle-counting infrastructure at the locations of the noise sensors. The analysis is restricted to sparse traffic conditions in this initial study.

State-of-the-art urban road traffic noise propagation simulation methods such as the CNOSSOS-EU framework rely on ray tracing to estimate noise levels at specific locations on façades, so-called receiver points; this method is relatively computationally expensive and its cost increases with the number of receiver points, which limit the spatial resolution of such simulations in the context of real-time or near-real-time urban noise simulation applications. This contribution aims to investigate the applicability of multiple data-driven methods to the surrogate modelling of traffic noise propagation for fast façade noise calculation as an alternative to these traditional, ray-tracing-based methods. The proposed approach uses compressed sensing to select a small subset of optimal receiver points from which the dataset of the entire façade may be reconstructed, associated with either a kriging model or neural networks, used to predict noise levels for these sensors. The prediction performance of each of these steps is evaluated on an academic test case, with two levels of complexity based on the dimensionality of the problem.

Road traffic is a major source of environmental noise pollution in urban areas. While strategic noise maps are widely used to identify the critical areas and propose mitigation plans, more specific tools are needed to evaluate the impact from traffic noise such as overall population exposure or anticipated impact from specific vehicles in varying spatiotemporal traffic conditions.

The present contribution proposes a receiver-centric mapping approach, introducing “noise-exposure sensitivity maps”, meant to assess the potential noise exposure impact from a specific vehicle in a given network, quantifying the associated exceedance over the prevailing background noise, under varying spatiotemporal traffic conditions. The resulting maps are thus focussed on a representation of the receiver exposure as opposed to considering the noise emission and propagation alone.

The complete methodology, its underlying assumptions, and possible applications such as route optimisation are demonstrated on realistic scenarios.

Responses to COVID-19 altered environmental exposures and health behaviours associated with noncommunicable diseases. We aimed to (1) quantify changes in nitrogen dioxide (NO2), noise, physical activity, and greenspace visits associated with COVID-19 policies in the spring of 2020 in Barcelona (Spain), Vienna (Austria), and Stockholm (Sweden), and (2) estimated the number of additional and prevented diagnoses of myocardial infarction (MI), stroke, depression, and anxiety based on these changes. We calculated differences in NO2, noise, physical activity, and greenspace visits between pre-pandemic (baseline) and pandemic (counterfactual) levels. With two counterfactual scenarios, we distinguished between Acute Period (March 15th - April 26th, 2020) and Deconfinement Period (May 2nd - June 30th, 2020) assuming counterfactual scenarios were extended for 12 months. Relative risks for each exposure difference were estimated with exposure-risk functions. In the Acute Period, reductions in NO2 (range of change from -16.9 mu g/m3 to -1.1 mu g/m3), noise (from -5 dB(A) to -2 dB(A)), physical activity (from -659 MET*min/wk to -183 MET*min/wk) and greenspace visits (from -20.2 h/m to 1.1 h/m) were largest in Barcelona and smallest in Stockholm. In the Deconfinement Period, NO2 (from -13.9 mu g/m3 to -3.1 mu g/m3), noise (from -3 dB(A) to -1 dB(A)), and physical activity levels (from -524 MET*min/wk to -83 MET*min/wk) remained below pre-pandemic levels in all cities. Greatest impacts were caused by physical activity reductions. If physical activity levels in Barcelona remained at Acute Period levels, increases in annual diagnoses for MI (mean: 572 (95% CI: 224, 943)), stroke (585 (6, 1156)), depression (7903 (5202, 10,936)), and anxiety (16,677 (926, 27,002)) would be anticipated. To decrease cardiovascular and mental health impacts, reductions in NO2 and noise from the first COVID-19 surge should be sustained, but without reducing physical activity. Focusing on cities' connectivity that promotes active transportation and reduces motor vehicle use assists in achieving this goal.

The Environmental Noise Directive (2002/49/EC) requires all European Union Member States to produce strategic noise maps using a common assessment methodology: CNOSSOS-EU. The reliability of CNOSSOS-EU railway noise evaluation is dependent on the input vehicle and track transfer functions. The CNOSSOS-EU default database contains the currently available choices for these transfer functions. However, these available transfer functions are limited and of insufficient quality, resulting in large errors in noise level calculations. An approach is presented, introducing an established analytical railway rolling noise calculation technique (TWINS), to extract more reliable and specific transfer functions. A case study consisting of railway rolling noise mitigation measures is defined and used as the basis for extracting and testing these transfer functions. The extracted transfer functions reduce the average deviation between CNOSSOS-EU and reference calculations using TWINS from6.1 dB(A) to 0.8 dB(A) in absolute sound power levels, and from 1.2 db(A) to 0.3 dB(A) in estimates of noise reduction potential for the defined mitigation measures. Application of this approach shows potential to improve the quality and depth of the existing CNOSSOS-EU default database. This may lead to more reliable estimations of railway noise in the strategic noise maps and the subsequent assessment of its harmful effects.

Improvements in noise mapping techniques and smart cities infrastructure have fostered the development of new ways to evaluate traffic noise exposure. An example of one such outcome is the noise exposure sensitivity map, which quantifies the noise exposure potential of a road network as a function of the vehicle type, the prevailing background noise, and the population exposed. The potential for planning vehicle routing that is offered by the above-mentioned map calls for solving the Vehicle Routing Problem (VRP) with a focus on optimising the noise exposure. A case study is chosen for applying the VRP, and it is taken from last-mile off-peak deliveries performed in Stockholm, Sweden in the context of the CIVITAS Eccentric project. The VRP is independently solved for the following objectives: distance travelled, driving time, and driving noise exposure potential. Also considered is a heterogeneous objective that is a combination of these factors. The impact of the objective function, on the resulting routes is presented.

Traffic is a major source of environmental noise pollution in urban areas. The present contribution focuses on a methodology designed to assess the impact of the noise generated by individual vehicles on a city's population using NoiseModelling, an open-source library implementing the CNOSSOS-EU model, capable of producing environmental noise maps. The initial step of the proposed method consists in processing microscopic traffic data (simulated in the present contribution), where the traffic intensity is dependent on the time of day that is targeted. The micro-traffic data is subsequently used to generate background noise maps by simulating the propagation of traffic noise. Then, the impact of the noise from the vehicle of interest is simulated, based on several parameters (route followed, type of motorization: diesel or hybrid, etc.). Finally, the data is post-processed to calculate the "exceedance" (increase in ambient noise) caused by the vehicle, taking the previously calculated background noise maps as reference. The complete methodology, its underlying assumptions, and the associated criteria proposed in order to assess the impact of noise emissions from individual vehicles is demonstrated on a realistic scenario.