Skip to main content
European Commission logo print header

QUIetening the Environment for a Sustainable Surface Transport

Final Report Summary - QUIESST (Quietening the Environment for a Sustainable Surface Transport)

Executive summary:

1 Executive summary

Context
European Commission addresses transport noise through 2002/49 (END): this directive promotes the reduction of the environmental noise. With expected noise reductions of about 10 to 20 dB, no action limited to a single step of the whole process could obtain such targets: one should act (and optimise the means of action) at all the consecutive steps of the whole process (sound - emission, propagation, and reception). Acting on sound propagation, ground transport Noise Reducing Devices (NRD) play an important role in reducing noise: depending on numerous different factors, their global effectiveness can drastically vary. Many efforts had already been done separately on the product side, and on the in-situ side of the NRD performances, while limited research had been done in order to integrate both sides: the final performance clearly depends on both in a true holistic approach.

Objectives
QUIESST merges, within a true holistic approach: the "true" intrinsic products performances, whatever their materials and shapes, together with their extrinsic ones, in order to assess their actual global capacities to reduce the amount of people exposed to noise (END target). It addresses: the near / far field relationship (linking the intrinsic characteristics to their corresponding extrinsic far field effects), the in-situ measurement methods of sound absorption (/reflection) and airborne sound insulation (methods relevant with the actual intended use, also allowing long term performances control), the 1st EU NRD database (listing and comparing both existing and new tests results and providing relevant relationships), the optimisation of the NRD's global performance through a holistic approach (considering acoustic, non-acoustic and global impact optimization, multicriteria optimization strategies, and possible global performance indicators), the sustainability of NRD (defining the relevant generic criteria and developing the first NRD's overall sustainability assessment method). QUIESST outcomes are integrated in the "Guidebook to NRD optimisation".

Work programme
WP1: project administrative, scientific and technical management.
WP2: definition of a far field effect indicator; development and validation of a numerical simulation method converting near field patterns to far field effects; development and validation of an engineering method for the translation of near field measurement data into far field reflection effects.
WP3: new measurement method for sound absorption (/reflection) and airborne sound insulation with regard to sound sources, signals, multiple sensors, signal analysis and the physical representativity; execution of a full inter laboratory (Round Robin) test to assess uncertainty.
WP4: collection and analyse of laboratory and in-situ tests results concerning sound absorption and airborne sound insulation; build-up of a comprehensive database of test results (different EU NRD), establishing the relationship between laboratory and in-situ measurements.
WP5: optimisation strategy for typical roads and railways (urban / rural); application to intrinsic and extrinsic performances, and to holistic optimisation (acoustic, non-acoustic and environmental), database of results from these optimisations; global impact of optimised solutions, case studies.
WP6: assessing the overall NRD's sustainability: defining relevant generic criteria (design, materials, construction technology and practice, maintenance, decommissioning...); establishing relevant assessment method; database of generic relevant criteria and indicators for EU NRD; case studies.
WP7: project dissemination, including the publication of the "Guidebook to NRD optimisation".

Results and achievements

WP1: successfully finalised project, relevant links with NRD stakeholders, special attention to CEN TC226/WG6 and TC256/SC1/WG40 working groups, ready to update standards or start new ones.
WP2: far field low- and high-rise buildings indexes (DLRIff,LR, DLRIff,HR), database of 1.200 NRD variants, derived and validated engineering method through a user-friendly public Excel sheet.
WP3: 2 new RRT-validated in-situ measurement methods with relevant uncertainty assessments, 2 new draft proposals for CEN 1793-5 and 6, as well as for CEN 16272-5 and 6, ready for WG analysis.
WP4: significant database of 1.421 test results on 414 EU NRD corresponding to 25 test laboratories and 9 countries, comprehensive easy-to-use public web database including relevant analysis tools.
WP5: intrinsic, extrinsic and holistic optimisation methods on acoustic, economic and environment factors, optimized NRD database with integrated tool (to be published), 3 global impact case studies.
WP6: assessment method using relevant generic sustainability criteria for NRD sustainability, 2 case studies being useful as models for the stakeholders to amend and tailor their own assessments.
WP7: http://www.QUIESST.eu relevant participation to major events, 39 papers, 6 publications, 2 workshops, publication of the "Guidebook to NRD optimisation".

Project Context and Objectives:

2 Project context and objectives
2.1 Background
If we think about how to ensure the sustainability of surface transport, then we definitely need to consider all the possibilities to reduce noise, as well as the sustainability of the associated devices used for noise reduction.

The European Commission clearly addresses transport noise through its 2002/49/EC Directive: its objective is to promote environmental noise reduction, and surface transport is one of the main targets. However, with EC expected impacts of noise reduction of about 10 to 20 dB, it is evident that no action limited to a single step of the whole noise problem could obtain such reduction in noise values: one should act (and optimise the means of action) at all the consecutive steps of the whole process (sound emission, sound propagation, and sound reception).

Acting on sound propagation, ground transport Noise Reducing Devices (NRD ) do play an important role in the reduction of noise: depending on numerous different factors, their global effectiveness could be as low as a few decibels (if used inadequately), or reaching up to 20 dB (while using appropriate design).Today, many efforts have been done on both sides of the characteristics leading NRD to be effective: the product side, and the in-situ side. However, too few and limited research has been done yet in order to integrate both sides, while the true final noise reduction clearly depends on both (in a true holistic approach).

The main idea of QUIESST is to optimise the knowledge, the methods, the use and the GLOBAL effectiveness of the ground transport NRD, in order to allow a durable and sustainable development of transport.

2.2 Overall concept and objectives
The global NRD performance depends on:
- the initial intrinsic acoustic characteristics of the industrial products used, and their sustainability;
- their relevant design (intrinsic acoustic performances, flat /non flat - homogeneous / heterogeneous devices, dimensions and location) in function of the vehicles, the infrastructure and the concerned environment;
- the whole sound propagation process: intrinsic performances which directly affect the near field propagation could affect the far field performances in a complete different way (remember that END can lead to more stringent noise reduction criteria, leading to more and more distant affected areas).

The concept of QUIESST is to merge, for the very first time, the consideration of the "true" intrinsic acoustic characteristics of NRD, together with their extrinsic acoustic characteristics, and their sustainability in a holistic way, in order to control the actual global effectiveness to reduce ground transport noise, to minimise the number of exposed people to noise and reduce the level of noise exposure and to make NRD more sustainable long term.

QUIESST aims to control all those important factors through a true holistic approach.

The main deliverable of QUIESST is a comprehensive reference guidebook about NRD holistic optimisation (referring to associated databases, simulation methods, measurement methods and recommendations: all these are also QUIESST deliverables).

QUIESST addresses: the near field / far field relationship, in-situ measurement of "true" sound absorption and airborne sound insulation, the comparison of the existing laboratory tests results of European NRD with the corresponding in-situ measurement test results, the holistic approach of NRD optimisation, and sustainability.

2.3 Detailed concept and objectives
To achieve its objectives, QUIESST clearly identified the following topics:

2.3.1 The near field / far field relationship
After more than 30 years using NRD alongside roads and railways, no definitive survey has been done yet in order to clearly demonstrate the global effect of specifically designed NRD.

In short, shape and sound absorptive materials are tools for achieving better noise reduction but, at present, it is impossible to properly simulate non flat and / or sound absorptive NRD effect in the far field: mastering the NRD performance, whatever their sound absorptive characteristics and / or shape in the near and in the far field is the QUIESST's 1st objective.

The main steps to achieve this objective were:
- to develop a numerical simulation method for the conversion of near-field sound reflection patterns to far field effects with NRD of different sound absorptions and / or shapes;
- to validate the numerical simulation method against measured data acquired in near and far field;
- to develop an (analytical) engineering computation method for the translation of near field measurement data into far field reflection effects to validate the engineering method against the results of the numerical simulation method and the available measurement data;
- to define an appropriate indicator for the rating of the NRD sound reflecting characteristics based on the far field effect.

The verifiable result is the validated engineering computation method, drafted with user friendly instructions for data processing and the corresponding far field indicator derivation.

2.3.2 In-situ measurement method of NRD intrinsic "true" sound absorption and airborne sound insulation
Since too long, one characterizes NRD intrinsic acoustic performances in close field and / or reverberant laboratories as if they were products to be used inside buildings (EN 1793-1 and EN 1793-2 ): this is inadequate relatively to their intended use, i.e. in open spaces. Moreover, this way does not allow an easy control of the NRD long term acoustic performances years after years, in facts a real need in order to assess NRD sustainability.

For in-situ measurements, the tentative CEN/TS 1793-5 is currently used by several Member States but has serious problems while characterizing / comparing flat and non flat products: as it stands, CEN/TS 1793-5 has been rejected as an harmonised EN standard.

Today, the need to characterize NRD in-situ is more than ever a priority if one wishes to master the NRD "true" intrinsic characteristics: addressing this is the QUIESST's 2nd objective.


The main steps to achieve this objective were:
- to develop a new measurement method for sound absorption/reflection and airborne sound insulation of NRD with regard to: choice of sound sources and signals, use of multiple sensors, signal analysis and the essential physical representativity (near field/far field, whatever the shape of the NRD);
- to assess the uncertainty of this new method (through a full Round Robin Test –RRT).

The verifiable results are the 2 new measurement methods and their uncertainty (assessment of accuracy).

2.3.3 Comparison between the laboratory and the corresponding in-situ tests results of existing NRD
The EU NRD market offers many already approved products (often tested under different methods), while many new ones are appearing. However, even if the European product standard EN 14388 is published since 2005, no comprehensive database of the NRD acoustic performances does exist yet. On the other hand, facing the expected coexistence of laboratory and in-situ tests results, the stakeholders strongly need to understand the possible relationships, if existing between in-situ test results and existing laboratory results.

Addressing both needs, the QUIESST 3rd objective was to build a relevant database comparing the European NRD intrinsic performances according to the different test methods, and to establish the relationships between the different results.

The main steps to achieve this objective were:
- to collect and analyze laboratory and in-situ tests results concerning sound absorption and airborne sound insulation (EN 1793-1/2, TS 1793-5, and the new QUIESST methods);
- to build a comprehensive database of test results, taking into account different EU NRD;
- to establish the relationships between laboratory and in-situ results and to supply data for a fair comparison of the two methods in terms of applicability;

The verifiable result is the database itself: a very significant one, including 1.421 test results on 414 EU NRD and corresponding to 25 test laboratories and 9 countries, this is presented under a comprehensive, easy-to-use, public web database including relevant analysis tools.

2.3.4 The holistic approach of how to optimise the use of NRD

Whatever the numerous existing "comprehensive" guides about NRD of these last 30 years, no one has yet included the holistic approach, i.e.: starting from the "true" intrinsic performances, considering the optimised combination of their acoustic characteristics and design shapes, considering the best situation in order not only to reduce noise, but also the amount of people exposed to noise, without forgetting the cost / benefit ratio and the sustainability...

QUIESST's 4th objective is to develop a comprehensive strategy on how to optimise NRD within a true holistic approach: this part of the project merges the results of the other parts (near/far field, "true" intrinsic performances, sustainability) together with all the other acoustic and non-acoustic considerations at global scale (road/rail, close/far field, urban/rural sites).

The main steps to achieve this objective were:
- to develop an optimisation strategy adapted to typical road and railway traffic noise configurations where both urban and rural areas are addressed;
- to apply this methodology to intrinsic performances, considering NRD shapes and surface impedances;
- to apply this methodology to extrinsic and holistic NRD optimisation, considering acoustic, non-acoustic and environmental (site) parameters, building a database of the results;
- to provide the expected global impact of optimised noise abatement solutions in terms of reduced number of exposed people in typical urban and rural situations (3 case studies: Belgium, the Netherlands and Germany);
- to merge the main outcomes from all the other parts of the project into a comprehensive final report of all the results issued in the project and giving recommendations and guidelines through good practices.

The verifiable result stands in the optimisation methods presented in the guidebook about NRD holistic optimisation (referring to associated databases, simulation methods, and measurement and assessment methods); moreover, a database of optimized NRD, with integrated tool helping the approach, will also be published.

2.3.5 Sustainability

Sustainability of surface transport is a key objective of the White Paper on European Transport Policy: it includes not only the vehicles and their infrastructure but also the numerous adverse effects they can have on the environment, noise being a major one. One then clearly understands the high interest to master all the systems which are able to reduce the number of affected people.

Optimised and sustainable NRD can play a very important part in this achievement towards a more sustainable ground transport. Furthermore, one also has to consider NRD as an integral part of the whole transport system, and their sustainability is equally important.

At present, there is no method allowing the assessment of NRD sustainability: QUIESST's 5th objective is to provide a relevant method for assessing the overall sustainability of ground transport noise reducing devices.

The main steps to achieve this objective were:
- to define the relevant generic sustainability criteria for NRD:
- sustainable design criteria, sustainable materials and their carbon footprint;
- sustainable construction technology and practice and their carbon footprint;
- sustainable maintenance; sustainable decommissioning;
- future sustainable solutions...
- to research relevant methods for assessing the overall sustainability of NRD;
- to build a database of those generic relevant criteria and indicators for existing European NRD;
- to apply the method(s) on existing NRD in order to compare and rank them from the point of view of their overall sustainability: this has been done through 2 cases studies (Italy and Spain);
- to present the methods and outcomes within the final report on NRD sustainability.

The verifiable result is the comprehensive report about NRD sustainability (referring to relevant parameters and generic sustainability criteria and associated assessment method): this report is also presented in the "Guidebook to NRD optimisation".

2.3.6 Dissemination
All the project results and outcomes have been distributed in the most transparent and effective way. Dissemination was a major part of QUIESST project as it ensured that the objectives and results of the projects were brought to the attention of targeted groups through appropriate dissemination channels.

The main steps to achieve this objective were:
- to exploit as much as possible the project's potential through a comprehensive review of previous and existing initiatives and potential target groups as well as a continuous clustering effort with all interested parties;
- to ensure that the objectives and results of the project are brought to the attention of these groups through appropriate dissemination channels (web site, articles and trainings);
- to confront the QUIESST expectations and conclusions with the needs expressed by the end users through dedicated workshops and by participating in major European and international events dealing with noise issue.

The verifiable results are: the QUIESST website (see http://www.quiesst.eu online), the relevant participation to major events, 39 papers, 6 publications, 2 workshops, and the publication of the "Guidebook to NRD optimisation".

Project Results:

3 Main Scientific and Technical results / foregrounds
3.1 WP2 "Near field - far field" relationship for sound reflectivity
Shape and sound absorptive materials are tools for achieving better noise reduction but, at present, it is impossible to properly simulate non-flat and / or sound absorptive NRD effects in the far field: mastering the NRD performance, whatever their sound absorptive characteristics and / or shape in the near and in the far field is the QUIESST's 1st objective.

3.1.1 The engineering extrapolation method: the final WP2's outcome

The method uses, as inputs, the results of the new WP 3 near field reflection test method: the 3rd octave band values of the averaged Reflection Index (RInf) are used. The barrier type and the geometrical shape parameters are also relevant inputs.

The output is an estimated contribution of the reflected sound to the sound level in the far field, expressed as the single number rating for the far field reflection index: DLRI,ff.

Locations that are considered for the sound source, the NRD and the receivers definitions of DLRI,ff,HR and DLRI,ff,LR

This single number rating, expressed in dB(A), is computed at five different receiver positions: at a distance of 100 m from the NRD, and at heights of 1.5 5, 10, 20 and 40 m above the ground.

The far field reflection index RIff is defined as the ratio between the amount of energy which is reflected by the device and the energy that would be reflected by a reference barrier (as a reference, a flat rigid vertical barrier of the same height as the test sample -usually 4 m - is chosen).

In order to obtain a compact description of the reflection effects in the far field the single number ratings at the five positions are then clustered and averaged in two groups: the average of the single number ratings of the three lowest positions DLRI,ff.LR is considered to be representative for low rise buildings and the average of the single number ratings of the highest two DLRI,ff,HR is considered representative for high rise buildings.

In this way, those two far field indicators characterise the far field reflectivity of NRD.

3.1.2 Basis of the engineering extrapolation method

The basis for the method is formed by the use of two data bases filled with results of numerical simulations.

Near field data base
The first database consists of results of simulations under the near field reflection tests conditions for different NRD variants representing the majority of the European NRD market.

Five different NRD families were selected.
Barrier category
Flat - tilted
Panes
Sawtooth
Zigzag
Steps

For each NRD type, 3 different types of absorptive material were applied:
1. Rigid: all materials with an acoustically hard surface (100 % reflective; 1 variant)
2. Porous concrete (6 variants)
3. Perforated metallic or plastic cassettes filled with mineral wool (6 variants)

The total number of variants in the near field data base is 1196. For each variant, the spectral values of RInf (near field Reflection Index) and the corresponding single number rating DLRI nf, averaged over three receiver positions are stored in combination with the material and geometrical parameter values.

Far field database
The second database contains the results of Boundary Element Model (BEM) simulations of the far field reflection index RIff values, for the same series of NRD variants as for the near field data. In this case, the values were computed for the five different receiver positions in the far field. For each receiving position, the far field single number indicators DLRI,ff have been also computed.

Step-wise extrapolation
The extrapolation is carried out in a two-step approach:
1. the result of a near field reflection test is matched to the best fitting simulated variant in the database, following a 2 steps matching procedure;
2. then, the material parameters (type of absorption material, flow resistivity and porous layer thickness) are used as input data for the computation of an estimate of the far field effects of the NRD: this estimate is computed with a polynomial approximation of the contents of the far field database. This enables a fast computation with the possibility to interpolate between the simulated variants.

The geometrical shape parameters are also used as input and these values can be interpolated between the values of the originally simulated variants in the database.

The final outputs of this far field extrapolation method are the two far field indicators DLRI,ff,LR and DLRI,ff,HR.

3.1.3 Uncertainty of the method
The engineering extrapolation method is a heuristic method, based on an approximation of the data that were computed with numerical simulation models for 1196 barrier variants: the approximations can deviate, to a certain extent, from the original simulated data.

Within the first step of the approximation process, the matching of the near field test results to the best fitting simulated variant was tested against the results of the WP 3 Round Robin Test: the differences between the single number ratings of the tests and the single number ratings of the fitted variants were always smaller than 1 dB, except for one very unusual design.

The second step estimates the far field reflection contribution for the best fitting simulated variant. It uses the material parameters of this best fitting variant and the barrier type and geometrical shape data. The basis of this estimation is a polynomial approximation of the far field simulation results that were computed with the BEM model. The estimated values have been compared with the original simulated values for all 1196 barrier variants and the 5 receiver positions.

Results of polynomial approximation Results of polynomial approximation sorted in order of geometric variation sorted in order of DLRI,ff

In this assessment of the estimation uncertainty the far field effects simulated with the BEM model are considered to be the "true" values: based on experiences in other studies there is a well-founded confidence in the reliability of the BEM simulation method, if it is used for modelling of sound propagation over relatively short distances.

The engineering extrapolation method derived from the BEM simulation results is presented with confidence and the uncertainty values specified above are seen as realistic estimates.

3.1.4 Examples of far field reflection effects computed with the engineering method

As an example, the data of the samples used for the WP3 Round Robin Test have been used here as input to the engineering method: both steps of the method (near field matching and far field extrapolation) were applied.

From those results, it can be seen that the far field effect does not always follow closely the near field reflection index values. This is logic and expected: if the barrier sample has a surface shape with large dimensions in vertical and horizontal directions, the far field effects of this surface design may be substantial and can enhance the reduction of reflections due to the absorption characteristics of the material.

In many cases the surface shape effects are also dependent of the receiver height.

3.1.5 Scope and availability of the engineering extrapolation method
The goal of the method is to give an indication of the far field reflection effects that can be achieved with a specific NRD design.

The scope of the method is limited to the NRD types and geometries considered in the database: if a specific design does not fall within that range, it cannot be assessed with the engineering method and new BEM simulations have to be carried out in order to obtain a reliable estimate of its reflectivity effects. The execution of a dedicated BEM simulation is also advisable if an assessment of the far field effects of a specific barrier design with less uncertainty is targeted.

The complete extrapolation method is described in a separate document in the format of a draft for an informative annex to the future revised standard for in situ testing of the reflectivity of noise barriers (EN 1793-5). For an easy use of the method, it is also implemented in a pre-programmed Excel spread sheet that is available to public through the QUIESST website (see http://www.quiesst.eu online).

3.2 WP3 Improvement of the in-situ methods for sound absorption and insulation measurement

3.2.1 Objectives of the new methods
The objectives were:
1. to develop new robust in situ measurement methods in order to assess the sound absorption/reflection and airborne sound insulation characteristics of NRD,
2. to assess the accuracy of those new methods.

The first objective implied that the new methods must be applicable on the site where the NRD are installed, without removing or altering them in any way and in presence of an unpredictable background noise, variations of meteorological conditions, traffic flows, etc. It should be kept in mind that the new methods are not intended to qualify NRD to be installed in almost "diffuse sound field" conditions, e.g. inside tunnels or deep trenches: in those cases, the traditional laboratory methods supply the necessary information.

The second objective has been achieved by assessing the so called "uncertainty" of the measurements by means of an inter-laboratory test (or Round Robin Test, RRT). In this context, the word "uncertainty" means a quantitative evaluation of the reliability of the results; it should be noted that it doesn't mean "error" or "wrong result": on the contrary, the declaration of the uncertainty is the best way, according to the recommendations of all international standard organizations (ISO, CEN, OIML, etc.), to assess the accuracy of a measurement.

Some more technical data are given in 3.5.3; the full description of the inter-laboratory test and its outcomes are given in the QUIESST deliverable D3.5.

3.2.2 Outline of the new in situ measurement method
In situ sound reflection measurement
An artificial sound source (loudspeaker) and a square array of 9 microphones (0,80 x 0,80 m) are used. Multichannel acquisition can be exploited. The array is placed between the loudspeaker and the device under test. The sound source emits transient sound waves that travel through the microphone array to the device and then reflects on it.

The microphones receive both the direct sound travelling from the sound source to the device under test and the reflected sound (including scattering).

A free-field measurement, taken for each microphone with the same source and microphone configuration but far away from any reflecting object, is then subtracted from the previous one in order to isolate the reflected component.

Several technical improvements (specifications for analysis windows application, a new algorithm for signal subtraction, a quantitative criterion for measuring the quality of the subtraction, etc.) have been developed in order to assure accurate results, even in difficult conditions.

From the ratio of the acoustic power of the direct and the reflected components, averaged on the nine microphones, a characteristic quantity is calculated: the sound reflection index RI. It is a dimensionless quantity, presented as a function of frequency in the 3rd octave bands between 100 Hz and 5 kHz. From those frequency dependent values, a single-number rating can be calculated, called DLRI and expressed in decibels.

In this formulation three newly defined "corrective factors" are included to master all the details of the measurement: a geometrical divergence correction factor taking into account the path length difference between the direct and reflected waves, a directivity correction factor taking into account the sound source directivity, and a gain correction factor used to compensate any gain mismatch (if any) of the amplification settings between the "free-field" and "barrier" measurements.

All this gives RI values physically meaningful and independent of the sound source used.

In situ airborne sound insulation measurement
The sound source emits a transient sound wave that travels toward the device under test and is: partly reflected, partly transmitted and partly diffracted by it. The microphone array placed on the other side of the device under test receives both the transmitted sound pressure wave travelling from the sound source through the device under test, and the sound pressure wave diffracted by the top edge of the device under test.

If the measurement is repeated without the device under test between the loudspeaker and the microphone, the direct free-field wave can be acquired.

From the ratio of the acoustic power of the direct and the transmitted components, energetically averaged on the nine microphones, a characteristic quantity is calculated: the sound insulation index SI. It is a dimensionless quantity, expressed in dB and presented as a function of frequency in the 3rd octave bands between 100 Hz and 5 kHz. From the frequency dependent values a single-number rating can be calculated, called DLSI and expressed in decibels.

Repeatability and reproducibility
The above outlined methods have been verified by 8 independent laboratories on 13 samples installed on 2 test sites in Grenoble and Valladolid (Spain).

Overall, the test has been carried out following the procedure for an inter-laboratory test

(also called Round Robin Test, or RRT) in order to be able to get both the repeatability and the reproducibility of the method.

The repeatability is the random variation of the measurement result under constant measurement conditions, while the reproducibility is the random variation of the measurement result under changed conditions of measurement.

The reproducibility is directly used to declare the reliability of the method according to the ISO guide on uncertainty in measurement. In other words, if M is the value of a single measurement and R is its reproducibility, there is a probability of 95% that the true value of a single measurement lies in the interval [M – R; M + R].

It is worth noting that these results have been achieved on real-life samples, built as in practice with irregularities and sound leaks due to average workmanship; in other words, these samples were not fully homogenous "laboratory samples". Thus, the final repeatability and reproducibility values do include the effect of sample irregularities.

In this regard, the final values obtained, already satisfying as they are, may be considered a worst-case estimate.

Thick red line: median value, Light red area: range between min. and max. values

Table of the 95% credible intervals for reproducibility and repeatability of the single-number rating of the sound reflection index DLRI in dB

Thick green/blue lines: median value, Light green/blue areas: range between min. and max. Tables of the 95% credible intervals for reproducibility and repeatability of the single-number rating of the sound insulation index for the acoustic elements and at posts DLSI in dB

3.3 WP4 Database of Acoustic performance of the European NRD
3.3.1 Overview of the Database Content
The NRD database contains data obtained with the different methods presenting single number rating and 3rd octave band spectra for different NRD families. The database contains 414 different NRD produced by 40 noise barrier manufacturers, and more than 1421 different measurement results, from tests performed by 25 different laboratories from 9 European countries.

More than 400 test results are on in-situ sound reflection, around 120 are on sound absorption measured in the laboratory, while 250 test results are concerning in-situ sound insulation and 100 sound insulation measured in the laboratory.

The measurement methods currently covered in the database are the following:
- Laboratory measurements for sound absorption and sound insulation according to
EN 1793-1 and EN 1793-2,
- The so-called "Adrienne" in-situ method for sound absorption and airborne sound insulation according to CEN/TS 1793-5, and prEN 1793-6
- The newly developed QUIESST method for measurements of sound reflection,
- French in-situ method for sound absorption and airborne sound insulation
NFS 31089.

The collected data represents the EU market distribution fairly well: most of the available data come from wood-fibre concrete, metallic cassettes and timber barriers, while transparent materials, photovoltaic barriers, added devices and green walls are less represented in the database.

3.3.2 The "internal" database
The database serves two different objectives: the first one is to perform an in depth statistical analysis of the current and historic data, the second one is to provide information about NRD for the general public, and especially for road and railways administrations. However, this leads to the major issue of confidentiality. On the one hand, there was a need to collect as many data from the manufacturers as possible to perform the analysis, while not all manufacturers and research institutions want to share this detailed information with the public and especially their competitors. For this reason, the detailed content of the so-called internal database will not be accessible to the public. Because of this, a second version of the database has been developed using only anonymous data and statistical information about the different NRD classes. Infrastructure administrations can check the currently possible performance of different classes while the manufacturers test reports and confidential information will not be publically available.

3.3.3 Examples of case Studies from the "internal" database

The lower right plot presents sound insulations results at post and at the acoustic element.

Timber barrier:

The frequency in the remark field indicates the used lower cut-off frequency for the test

3.3.4 The "public" database

The public database is the main output of WP4: it is directly accessible for all the stakeholders from the QUIESST website. The public database is based on the analyses performed with all the data collected during the project. For confidentiality reasons only an overview of the data and the results of the analyses can be presented to the public and not the data itself, which are present only in the internal database. Here are some "menus":

In addition to the overview of the single number ratings and the frequency spectra, more detailed analyses and comparisons between the different methods have been performed. In this section, the following analyses and comparisons are presented:
- Correlation between laboratory and in-situ method for sound insulation over all barrier types (EN 1793-2 and prEN 1793-6)
- Correlation between laboratory and in-situ method for sound insulation for each material where sufficient data were available (EN 1793-2 and prEN 1793-6)
- Correlation between laboratory and in-situ method for sound absorption/reflection over all barrier types (EN 1793-1 and CEN/TS 1793-5)
- Correlation between laboratory and in-situ method for sound absorption/reflection for each material where sufficient data were available (EN 1793-1 and CEN/TS 1793-5)
- Comparison between in-situ sound insulation measurements performed in front of a post and measurements performed in front of a noise barrier element over all barrier types (prEN 1793-6)
- Comparison between in-situ sound insulation measurements performed in front of a post and measurements performed in front of a noise barrier element for each material where sufficient data were available (prEN 1793-6)
- Comparison between different methods for in-situ sound reflection, where sufficient data were available (CEN/TS 1793-5 and NFS 31089)
- Comparison between different methods for in-situ sound insulation, where sufficient data were available (prEN 1793-6 and NFS 31089)
- Cluster analysis of the collected results for each measurement method separately in order to identify NRD families based on the frequency spectra

Based on the huge amount of data collected, it will be possible to perform many other analyses in follow-up research.

3.4 WP5 Holistic optimisation of NRD
The main challenge of this Work Package was to develop an original optimisation methodology dedicated to complex shape NRD, taking into account acoustic and non-acoustic parameters simultaneously through global performance indicators.

The goal was not to produce the "best optimised NRD", but instead to give the opportunity to engineers as well as manufacturers to re-use this approach for their own research.

3.4.1 Choice of best practice models
Sound propagation models
Four main 2D sound prediction models were selected to be the most pertinent for the purpose of accurately predicting complex shape NRD performances:
- The BEM (Boundary Element Method), very flexible to model noise barriers of complex shape including impedance jumps and curved surfaces. On the other hand, BEM ignores the effects of atmospheric gradients due to meteorological effects and should be used for predictions not too far from the NRD (100 m propagation max);
- The FDTD (Finite-Difference Time-Domain) model considering atmospheric refraction and therefore useful to include meteorological conditions in the optimization of barrier shapes. However, FDTD is a bit less flexible than BEM for modelling complex shapes;
- The TLM (Transfer Line Matrix) offering flexibility in the description of the geometry of the boundaries with atmospheric refraction taken into account;
- The TMM (Transfer Matrix Method) dedicated to the prediction of sound transmission and absorption through a multi-layered noise barrier.
We also suggest using hybrid models such as the FDTD-PE and BEM-PE (PE for Parabolic Equation model) for NRD effects at long ranges taking into account meteorological effects.

A 3D asymptotic model such as the Ray method is recommended when studying the global impact of NRD on realistic large built areas (see 3.4.4 hereafter).Then the model should be adapted to complex situations by including results from BEM, FDTD, TLM and TMM.

Optimisation models
As regards with selection of best optimisation models, our recommendations are:
- Concerning mono-objective optimizations the evolutionary strategy is the most relevant, since many parameters have to be simultaneously optimized;
- Concerning multiple-objective optimization, both approaches by aggregated methods and Pareto methods are advised;
- The construction feasibility of the optimized NRD should be taken into account in order to avoid unfeasible noise abatement solutions.

3.4.2 Acoustic and non-acoustic optimisation indicators
Acoustic indicators
Intrinsic optimisation means that one evaluates any acoustic performance in the vicinity of the noise barrier, ignoring its own environment and considering a point noise source. The performance indicators we used were those calculated in the relevant EN 1793 standards: the reflection index DLRI, the transmission index DLSI and the diffraction index DL-DI.

Extrinsic optimisations are achieved considering the NRD in its environment: real sound sources, infrastructure heights, topography and, eventually, buildings. We calculated the sound level difference IL as the acoustic indicator, for receivers located on both sides of the infrastructure. IL represents the acoustic gain obtained with an optimized NRD compared to the reference concrete barrier.

Environmental indicators
As a result of a specific Life-Cycle Assessment (LCA), a set of environmental indicators was proposed. Among them we recommend to utilize the four ones used in QUIESST: Energy, GWP (Global Warming Potential), Waste (non-hazardous and inert) and Water consumption. These environmental indicators were evaluated for a set of 8 common materials used in NRD engineering (wood concrete, timber...) on a basis of a common functional unit, chosen here to be the production of 1000 kg of material and its transport over 100 km. We also took into account the reference service life of each material exploited. Recommended values are available. We finally used the ratio of the indicator value to the one of the reference barrier.

Cost indicators
In our approach, the cost indicator was the sum of three parameters: construction, maintenance and demolition costs. Demolition costs included transportation but did not consider material re-use. Applicable values have been proposed. As previously, we used the ratio of the indicator value to the one of the reference barrier.

3.4.3 Holistic optimisation methodology
Description of the methodology

Starting with the random creation of a set of 50 different NRD within fixed NRD family and environmental situation (source/area/topography), an evaluation of the acoustic, environmental and cost indicators is achieved, and a linear averaging is done to obtain 3 aggregated indicators: ACOU, ENV and COST. All these indicators are compared to those obtained for the reference NRD: a straight, rigid concrete barrier. Then 12 new NRD (25% of 50) are created with limited changes (in shape and material) from the 12 "best" NRD they finally replace. Hence a new set of 50 NRD is revaluated. This process ends when the highest values of all indicators vary by less than 5% from an evaluation step to another.

Application to typical NRD families
This holistic optimisation methodology has been applied to acoustic and non-acoustic performances of 4 generic NRD families in different environmental situations including road and railway sources, rural (absorbing ground) and urban (rigid ground) areas, as well as flat, embanked (+5 m) and depressed (-5 m) topographies.

A grading system [12] has been applied to the 3 aggregated indicators in order to express them on a dimensionless scale ranging from 0 (bad) to 4 (very good). A radar plot display is recommended to present these 3 global NRD performance indicators.

Optimized NRD database
All extrinsic NRD optimisation results were recorded in a database that can be queried through a simple tool. The 1st step consists in selecting the type of source as well as the environmental configuration, the infrastructure topography and the NRD family.

One also may select one of 3 following configurations for calculation of the ACOU indicator: receiver at the source side only (sound reflection), receiver at the receiver side only (sound diffraction), or receivers on both sides.

Then, the user has to select one optimized solution among the set of final optimized NRD obtained at the end of the optimization process. To do so, the user should tune the three aggregated indicators ACOU, ENV and COST to the desired weights (in percentage), 0%, 50% and 100% meaning minimum, medium and highest importance, respectively.

In order to select the solution corresponding to the best ACOU aggregated indicator (whatever the ENV and COST indicators), one has to tune as follows: ACOU=100, ENV=0, COST=0. One can also display all results one by one, using the function "individuals".

Finally, the selected optimized solution is displayed, giving:
- The general shape of the optimized NRD in a vertical section,
- Materials used and their location on the NRD surface,
- The NRD optimization shape parameters (width, tilting, roughness size, etc.)
- The corresponding values of ACOU, ENV and COST

Example of use of the Optimized NRD database
An example of typical results one can get from the database is presented hereafter.

Considering the case of a strongly non-flat barrier along a motorway on a flat, rural terrain with the receiver at the source side only (sound reflection), we extracted three solutions optimised in priority for ACOU, ENV and COST, whatever the values of the two other aggregated indicators: one may note the great diversity in shapes and materials used depending of the choice of the indicators' weightings.

The database could be re-used and adapted at the upstream phase of future traffic noise impact projects in order to globally assess the potential acoustic gain that may be obtained by optimising (in shape and in material type) a conventional noise barrier taking also into account both the environmental impacts and the cost efficiency.

3.4.4 Global impact
We also aimed at using the previously optimized NRD and placing them in a realistic 3D built environments. With the use of a sophisticated multidimensional interpolation model and a ray tracing method OASIS developed at CSTB, we showed the ability to determine how much these optimized NRD could reduce the amount of people exposed to high noise levels.

Application
3 different types of dwellings were considered: collective (21 mH), semi-collective (9 mH) and individual. Different optimised NRD were tested. Final results were given through histograms showing for each of the studied cases (depending on type of optimized barrier, type of dwelling, road infrastructure) the proportion of inhabitants subject to a sound level abatement (?L) by step of 1 dB(A). In this approach, we distinguished people living at lower, intermediate and upper floors.

We also calculated the population exposure indicator difference Lden,pop that represents the difference between the Lden, pop obtained with the reference NRD and the Lden,pop obtained with the optimised NRD in terms of global sum of noise level of all residents on the most exposed facades: values of L ranged from 0 to 8 dB(A), when average values on all receivers Lden,pop were from 0 to 5 dB. The highest values of L were obtained for the lower (ground and first) floors of the semi-collective housing, the 2nd and 3rd floors of the individual houses, and the upper floors of the collective housing.

Another way of presenting global results is to give for a specific case the proportion of population benefiting from a noise abatement of at least 3 dB(A): in this research, depending upon the type of optimised barrier and type of dwellings considered, the proportion varied from 1% to 70%, pointing out that NRD optimization should be realised for very specific noise situations.

3.5 Sustainability
Assessing sustainability involves measuring and evaluating many and conflicting attributes in an unbiased way. In order to assist the relevant stakeholders to assess the sustainability of NRD projects with the view to complying with and supporting the transport and overall global sustainability agenda, the following key novel QUIESST outcomes are presented hereafter.

3.5.1 Defining 'Sustainability' for NRD
NRD sustainability has been defined as the following: 'The optimal consideration of technical, environmental, economic and social factors during the design, construction, maintenance and repair, and removal/demolition stages of NRD projects'.

3.5.2 Sustainability Key Performance Indicators for NRD projects
The Sustainability Key Performance Indicators (KPIs) are essential components in the overall assessment of progress towards sustainable development. They are useful for monitoring and measuring the sustainability state of a NRD project by considering a manageable number of variables considered critical to sustainability.

KPI n° per Sustainability Factor NRD Sustainability Assessment Criteria Key Performance Indicator
(possible way of measurement) Benchmark to Improve Sustainability Performance
Social Acoustic comfort No. of complaints from residents Reduce
Work related sicknesses and Injuries No. of reported health incidents/work related injuries due to working conditions Reduce
Vulnerability of the barrier to vandalism No. of reported acts of vandalism to the NRD (includes graffiti) Reduce
Glare control for road users No. of reported road accidents due to the glare from the noise barrier to the emergency services Reduce
Crossing facilities such as footbridges/ underpasses No. of complaints from the impacted community due to the lack of adequate crossing facilities Reduce
Acceptance of the architectural design of the NRD No. of complaints due to the architectural design of the NRD Reduce
Loss of view for residents and road users No. of complaints from residents and road users due to loss of views Reduce
Barrier design/type via public consultation No. of projects that included (and implemented) a stakeholder engagement plan Increase
Use of local companies and labour No. of local companies employed/No. of local labour opportunities realised Increase
Social acceptability of the NRD No. of complaints from residents Reduce
Technical Use of new materials % new(virgin)material content/m3 or m² or m Reduce
Use of recycled materials % recycled material content/m3 or m² or m Increase
Local materials % local material content/m3 or m² or m Increase
Whole barrier service life Years Increase or maintain
Acoustic durability in-situ years (yrs.) until acoustic performance drops below the accepted level Increase or maintain
Buildability/constructability of the noise barrier square meter/day to build the noise barrier system Increase
Durability No. of years the NRD system can be used in comparison to its design life Increase
Environmental Loss of land 'Footprint' (m²) of the NRD/m or total length Reduce
Overall waste production kg/m² Reduce
Materials used for energy recovery at the end of its life % material recoverable for energy/m² Increase
Recyclability potential % recyclable /m² Increase
Re-use potential % re-usable/m² Increase
Carbon footprint (global warning potential) kg CO2equivalent/m² Reduce
Water footprint litre/m² Reduce
Embodied energy content (Use of primary energy resources/consumption) MJ/m² Reduce
Renewable energy production (Photovoltaic/small scale wind turbines) MJ/m² Increase
Economic Capital costs Euro/ m² Reduce
Maintenance and repair costs Euro/ m² Reduce
Removal/replacement costs Euro/ m² Reduce
Income generation Euro/ m² Increase

3.5.3 Relevant Generic Sustainability Criteria for Assessing the Sustainability of NRD
Sustainability criteria highlight issues that are important for sustainability assessment. Primary criteria are not usually measurable, and will typically have a set of secondary criteria which define the primary criteria. Secondary criteria underpin the primary. These are measured through the use of indicators that are the 'Unit of measurement' for secondary criteria which may be either quantitative or qualitative. In some cases, secondary criteria may have further attributes/tertiary criteria that define them further and are measured through the use indicators, too.

A 'Top-Down-Bottom-Up (TDBU)' research strategy was developed and implemented to create and validate the relevant generic set of sustainability criteria for NRD projects. This mainly involved gathering expert opinion from the relevant stakeholders through a series of workshops, questionnaires and interviews. These '22 primary criteria' highlight all the major issues to consider, and assess across each sustainability factor. In total, 141 criteria form the complete sustainability hierarchy for NRD, of which, 92 are directly measurable.

Sustainability Factor Primary Criteria
Technical -Material selection
-Buildability/constructability
-Flexibility and adaptability
Environmental -Energy
-Land use
-Air quality and climate change
-Flora and fauna
-Water
-Waste
Economic -Life cycle cost
-Green value
- Financial sources
-Compensation cost
-Affect on local residential/commercial property prices
-Contractual and procurement type
Social -Safety and security
-Health and well-being
-Severance/separation
-Social acceptance
-Architectural design and local context
-Community engagement
-Local employment and engagement with local business

3.5.4 Generic database of sustainability criteria per main NRD type
Using the generic set of NRD sustainability criteria previously established, research was carried out to generate and collect indicative sustainability criteria values for the 13 main NRD types. Results have been tabulated into a database so that the sustainability performances of the 13 NRD types can be viewed and compared to either:
1. benchmark the sustainability performance of a given NRD type with respect to the average/generic data provided in the database for the NRD type in question, or
2. use the generic data in place of collecting site/system specific data when it is considered impractical (or in some cases not necessary) to conduct the sustainability assessment, and so reduce significant analysis time and costs.

3.5.5 Stages for Assessing the Sustainability of NRD via Multi Criteria Analysis Approaches (MCA)
The assessment of the sustainability of NRD is a multi-criteria analysis (MCA) problem as it involves selecting and assessing multiple conflicting NRD sustainability criteria.

Any MCA tools are able to generate a sustainability assessment index score in the range 0 to 1 or -1 to 1 for potential design and build NRD solutions, or built and operating NRD projects, relative to either the set of alternatives considered, or to a user defined baseline.

3.5.6 An Example Analysis of Assessing the Sustainability of NRD
An example of assessing the sustainability of a given NRD type (a steel noise barrier) using a small set of selected criteria and generic NRD sustainability performance data is given below. It should be noted that the assessment of sustainability is always a relative concept.

There are principally two relative assessment approaches:
1. The sustainability assessment is relative to the set of alternatives (or options) being considered, or
2. The sustainability assessment is relative to an absolute state/user defined baseline.

3.5.7 Overall Benefits of the NRD Sustainability Research and Contribution to the State of the Art for the NRD Industry
NRD Industrial Associations have been directly involved in this research both at national and European level and relevant benefits are expected from a common approach in sustainability evaluation and assessment.

NRD manufacturers have always shown a great interest in sustainability assessment due to its construction products being developed due to the environmental need of reducing noise disturbance in residential areas.

Potential Impact:

4 Impacts
4.1 Potential impacts
The first major impact of QUIESST will be the direct consideration of its outcomes at the coming meetings of the CEN TC226/WG6 (road NRD) and CEN TC256/SC1/WG40 (railway noise barriers): those meetings will be held on the 8th, 9th and 10th of April 2013 in Paris.

Those working groups are drafting the EU / CEN standards for the NRD products and use: they reply to the Construction Product Directive (CPD), the new Construction Product Regulation (CPR) and the Directive on Interoperability of transports.

Thanks to the (strongly expected) results of QUIESST, the following points will be considered by the working groups' experts:
1. submission of a new topic / new standard item on Far Field effects in the road and railway standards for NRD (WP2);
2. (re)drafting the EN1793-5 and 6 (roads) and corresponding EN 16272-5 and 6 (rail) with the new QUIESST methods in order to reach consensus toward true harmonized standards (WP3);
3. submission a new topic on Sustainability in the road and railway standards for NRD (WP6).

Of course, this will also be the perfect place to distribute the Guidebook to all the stakeholders present: manufacturers, authorities, road / railway companies, experts... The results of the 1st EU NRD database (WP4) and the optimization methodologies described in WP5 are also of main interest for all.

The standardization working groups are the best to value the outcomes of QUIESST: this will be assured by Jean-Pierre Clairbois, who is the convenor of TC226/WG6, and Massimo Garai, the TC256/SC1/WG40's convenor.

4.1.1 Holistic noise abatement
Acting on sound propagation, NRD's global performances can vary from a few decibels (if use in an inappropriate manner – what is unfortunately too often the case), or reaching up to 20 dB (while using appropriate design, materials and infrastructure integration). Facing this huge scale of performances, one clearly understands the QUIESST objectives, that is to optimize NRD global performance thanks to: product characterization relevant of the actual intended use, understanding how the relevant "true" intrinsic performances can act in the far field, a better and fairly balanced knowledge of EU NRD product market, and optimization strategies that really consider noise propagation in a holistic approach.

4.1.2 Reducing the amount of people exposed to noise

Reducing environmental noise of ground transport by 10 to 20 dB is a challenging objective; however, it is meaningless if it does not correspond to a relevant reduction of the amount of people exposed to noise, which is the target objective of the 2002/49/EC END Directive.

QUIESST not only addresses a more relevant characterization of NRD, whatever their materials, shape and design, but also the global NRD capability to reduce environmental noise at all the steps of the process, i.e.: from the vehicles to the finally exposed habited areas: by improving the knowledge about the near / far field relationship and by improving NRD with adapted method of characterizing their acoustic performances, a new NRD should give either a better noise reduction, or should be able to give similar performances to existing ones, but in such a way that some gain is found (lower height, design more integrated to the landscape, better sustainability)

4.1.3 Increasing NRD sustainability
Only optimizing noise reduction properties of NRD (without considering their sustainability and, in particular, their impact on the environment AND on natural resources and the social consequences that NRD can bring in the human environment) is the worst mistake done up to now.

More optimal NRD as a result of research carried out in WP2, 3, 4 and 5 will impact less on the environment and contribute less to climate change; it will be less costly as well.

On the other hand, the "intrinsic" NRD sustainability is a characteristic which was not yet defined, and needed to be. WP6 provides a relevant method for assessing the sustainability of NRD at all stages in the life of a NRD (design, construction, maintenance and repair and decommissioning), including their carbon footprint.

Through WP6, QUIESST contributes to the reduction of the impact on climate change when using NRD, and should reduce the use of natural resources through more sustainable materials and the usage of recycled ones.

4.1.4 Economic impacts
By delivering optimization strategies to all the relevant stakeholders, QUIESST helps to improve the costs / benefits ratio of NRD global performance, what will finally benefit to the whole Community.

On the other hand, QUIESST will have a very important impact on promoting European NRD Industry and also the European experts abroad. It could be relevant to mention how important it is that QUIESST helps European NRD Industry to improve their products design and promote them abroad.

4.1.5 Policy concerns

QUIESST replies to European policies requirements: the whole project clearly targets the achievement of the END Directive main objectives, i.e.: not only noise reduction, but also the reduced amount of people exposed to noise.

On the other hand, the project clearly addresses the Construction Product Directive (CPD) and the new Construction Product Regulation (CPR) as well as the Directivity on the Interoperability of transports., i.e.: the requirements to place products on the market only if they are fit for their intended use (WP2 to 4), and doing so in an economically reasonable working life (WP3 to 6), providing (methods) standards about essential characteristics that qualify products in a common, fair and relevant manner (WP 3).

4.2 Dissemination and exploitation of project results
In order to increase the awareness of research, industry, users and public authorities of the developed solution and the project in general, a broad range of dissemination activities has been provided in QUIESST (in line with the tasks envisaged as part of this strategy):
- Production of a Guidebook at the end of the project;
- A dedicated website;
- Leaflets distributed by the partners and diffused during relevant events;
- Contributions to technical literature, research journal papers, articles in newspapers (science and technology section);
- Organization of two open workshops for selected attendants (amongst which different stakeholders);
- Participation on international workshops and conferences and submission of abstracts if adequate.

List of Websites:

http://www.quiesst.eu