Journal of Environmental Management 220 (2018) 77–86

Contents lists available at ScienceDirect

Journal of Environmental Management

journal homepage: www.elsevier.com/locate/jenvman

Research article

Impact assessment of pollutants from waste-related operations as a feature T

of holistic logistic tool
Vlastimír Nevrlýa,∗, Radovan Šomplákb, Jiří Gregora, Martin Pavlasb, Jiří Jaromír Klemešb
a
Institute of Process Engineering, Faculty of Mechanical Engineering, Brno University of Technology – VUT Brno, Technická 2896/2, 616 69 Brno, Czech Republic
b
Sustainable Process Integration Laboratory – SPIL, NETME Centre, Faculty of Mechanical Engineering, Brno University of Technology – VUT Brno, Technická 2896/2,
616 69 Brno, Czech Republic

A R T I C LE I N FO A B S T R A C T

Keywords: Waste management has still been a developing and progressing field, which demands continual improvements in
Impact of pollutants waste transportation as well as proper selection of locations and technical operation of new treatment facilities.
Waste transportation Most of research papers on waste management planning have been dealing with optimisation of network flows,
Local environment assessment thus minimising the cost and improving economic criteria. The shortest paths to treatment facilities are con-
Network edge characterisation
sidered together with detailed analysis of their operation including heat and electricity demands in their vicinity.
The tasks sometimes include social and global environmental criterions, however, the direct local consequences
also play an important role and should be examined. A decision-making strategy in waste management updated
with the local emission impact on the population is proposed in this paper. The paper focuses on the first move in
analysing the production, dispersion, and impact of pollutants, originating in transport, with regards to the
population living close to routes. The calculation of emission produced during the transport of waste takes into
consideration the altitude profiles of routes, container loads, and specific types of vehicles. The consecutive
estimated impact on the population reckons with the distances between routes and municipalities as well as their
sizes in terms of the numbers of inhabitants, where the transportation routes are divided into smaller segments
and dispersion is limited with threshold value. The proposed approach describing the emission effect has been
tested using real-life operating data corresponding to the specific, 81 km long route along which approximately
25 t of waste is transported 800 times a year. The impact of pollutants on the population was evaluated and
discussed. Results of the analysis were quantified for this route to create an edge characterisation needed for
further calculations. This approach applied to the whole network then yields input data needed for future re-
search of novel strategies in facility location problems. Other possible extensions of the presented approach
include more accurate dispersion function or detailed calculation of the impact of pollutants with respect to
specific locations of residential houses.

1. Introduction
The impact of human activity on the environment becomes an in-
creasingly important issue. The paper by Olsthoorn et al. (2001) re-
views the existing literature on environmental performance indicators.
One of the frequently discussed issues within environmental manage-
ment is global warming (GW). Danny Harvey (1993) has proposed a
guide to global warming potentials (GWP) which summarizes the
comparison of different gases and their effects. As stated by Wirl
(2012), the future decisions should be strategic regarding prices and
quantities. One of the area with the important impact to GW is also
waste management. In (Zhang and Huang, 2014), the facility expansion
and waste flow allocation planning with consideration of greenhouse
gases (GHG) emission trading are analysed. Irrespective of whether the
CO2 equivalent is adopted as criteria to measure the impact, see (Čuček
et al., 2012) for GHG footprint analysis, all these indicators are con-
sidered globally.
On the other hand, many places around the world are facing the
problem of increasing air pollution, especially in the densely populated
areas such as large cities. This is primarily due to the centralisation of
industry in urban areas and major traffic loads; see (Cai et al., 2018) for
a comparison of carbon emissions from different sectors between urban
and non-urban areas. These are local issues, which have a direct impact
on the quality of lives (haze, frequent respiration problems) and, in the
long-term, increased rate of civilisation diseases. Transportation plays
an important role in the cities. The evaluation of different vehicle

∗
Corresponding author.
E-mail address: Vlastimir.Nevrly@vutbr.cz (V. Nevrlý).

https://doi.org/10.1016/j.jenvman.2018.05.003
Received 28 January 2018; Received in revised form 3 April 2018; Accepted 1 May 2018
Available online 12 May 2018
0301-4797/ © 2018 Elsevier Ltd. All rights reserved.
V. Nevrlý et al.
system technologies is proposed in (MacLean and Lave, 2003).
Waste management has still been a developing and progressing
field, which demands continual improvements in waste transportation
as well as proper selection of locations and efficient operation of new
treatment facilities. A comparative analysis of different assessments of
municipal waste management systems is reviewed in (Cleary, 2009) or
for food waste management systems in (Bernstad and Jansen, 2012). In
case of a new system design, the majority of research papers have been
dealing with optimisation of network flow to minimise the cost and
improve economic criteria, see e.g. (Goulart Coelho et al., 2017). Re-
garding mixed municipal waste (MMW), which is favoured for thermal
treatment, the shortest paths to facilities need to be considered together
with detailed analysis of operation including local demands for heat
and electricity. The tasks sometimes include social and global en-
vironmental criterions (GHG production), however, in case of emissions
it is not only their quantity which should be considered, but also their
local impact on a particular area with regard to the respective popu-
lation density. Thus, it is necessary to adapt decision-making which
interferes with local city environment towards the possible future
consequences.
This paper follows two objectives. First, the initial idea of building
advanced reverse logistic model is introduced. Potential locations of
waste treatment facilities are given and their optimal capacities with
regard to the treatment cost, transportation cost and production of
waste in the vicinity are optimised as in (Šomplák et al., 2014), how-
ever, the local emission impact assessment is included as well. The
approach aims at optimum waste treatment capacities allocation, while
future potential risk arising from transport through populated areas is
addressed.
An integral part of any reverse logistics problem is a model of in-
frastructure, which is a graph evaluated using multiple coefficients. Its
key elements are edges, each of which is assigned a value corresponding
to the locally based emissions. A model without the proper evaluation
would not bring any results. Thus, the second objective of this paper is
to propose a methodology for edge-related analysis, which provides a
deep insight into the problem. It can be considered as a pre-processing
phase carried out prior to the main calculation. An analogy to this with
capacity-dependent transport price as the function is mentioned in
(Gregor et al., 2017). The assessment of production and impact of CO2,
CO, NOx, and PM was carried out for a particular, 81 km long edge with
a specific (constant) amount of transported waste.
The paper is structured as follows: Section 2 starts with the state-of-
the-art analysis and then it summarizes idea of an advanced tool in the
field. An example of how the advanced tool can be used is described in
section 2.2. Together with overview of input data needed to solve real
case studies. An approach of handling point and line sources of emis-
sions and their dispersion is proposed in section 2.4. The approach is
evaluated using a small case study involving a single route in section 3.
2. Local assessment approach
2.1. Summary of the state-of-the-art and research challenges in the field
Many articles devoted to reducing emissions had considered other
environmental aspects as well, see (Cheng et al., 2017). These tasks
often combine economic and environmental criteria as done in (Yilmaz
et al., 2017). In this context, also Kanzian et al. (2013) introduced a
supply chain for processing of forest biomass. The task was designed as
a multi-criteria optimisation problem, where profit was maximised
while CO2 emissions were minimised. Further work focusing on second-
generation biodiesel supply chains was published in (Hombach et al.,
2016). The approach was, again, based on mixed integer programming
and the tool was presented on a case study from Germany. The results
contributed to the proposed emission reduction plan. The paper by
Havukainen et al. (2017) employed the environmental impact assess-
ment using an LCA program to determine whether refuse-derived fuel
78
Journal of Environmental Management 220 (2018) 77–86
production and incineration can have a more positive impact on the
environment than co-incineration of MMW with coal. An analysis
concerning carbon emissions was presented by He et al. (2016). The
work focused on a dual-channel supply chain which included tradi-
tional and online retailers. It evaluated the impact on pollution re-
duction as a result of changes in sales channels and the transport of
goods. Zohal and Soleimani (2016) presented a multi-criteria decision-
making problem, which minimises costs while reducing emissions. The
paper seeks to address how a multi-objective logistics model in the gold
industry can be created and solved through an efficient meta-heuristic
algorithm.
The previously mentioned articles mostly combine economic and
environmental problems from the global point of view (objective
function is minimised). More complex multi-objective location-routing
model for transportation of hazardous waste was developed and tested
in Turkey, see (Samanlioglu, 2013). Three criteria were minimised:
i. Total cost (transportation cost of hazardous materials and waste
residues and fixed cost of establishing treatment, disposal, and re-
cycling centres).
ii. Total transportation risk related to the population exposure along
transportation routes of hazardous materials and waste residues.
iii. The total risk for the population around treatment and disposal
centres.
The second and third items addressed the impact on the population
and Samanlioglu (2013) considered a certain number of people living in
a specific area or within a certain threshold distance from the routes.
The way to calculate the impact differs from task to task according to
the type of waste and its properties – hazardous waste becomes an issue
when the vehicle is involved in an accident. However, in case of MSW, a
detailed analysis should be carried out. It is important to consider the
distance from the emission source (vehicle, treatment facility), the
prevailing direction of wind, as well as the related pollution dispersion.
To the authors' knowledge, these aspects have not been addressed in
reverse logistics problems yet. The emission models and approaches
proposed in section 2 serve as a preparation phase (input data for the
calculation) for a comprehensive assessment of supply chains using the
facility location model with regard to the ideas presented by Bing et al.
(2016). Such a model should consider waste transportation (distance,
route profile, load capacity of a lorry) and allocation of treatment plants
(Waste-to-Energy (WtE), landfill sites, mechanical-biological treatment
producing alternative fuel for co-incineration, transfer stations) in the
decision-making process in a holistic approach where local impacts are
addressed in defining global solutions where high-level details are re-
quired.
2.2. An illustrative example of where an advanced tool is needed
Let us first present an actual case related to the Czech Republic,
where there are four WtE plants in operation. The largest one is situated
in the capital city of Prague. With its processing capacity of 310 kt/y, it
processes residual waste produced in the agglomeration with a popu-
lation of 1.3M inhabitants. Considering the overall residual waste
production of 400 kt of MMW in 2016, additional WtE capacities are
needed. As for the heat utilisation, the outmost location of the potential
WtE plant is 35 km away, which corresponds to the city of Mělník
(96,000 inhabitants). As an alternative, enlargement of the existing
Prague facility is also possible. The decision whether it is worth to build
a facility outside the populated area can only be made after additional
criteria except cost are considered. It is obvious that the impact of
additional emissions in Prague caused by the WtE plant would be
higher than in case of Mělník, but it is not the only constraint.
Fig. 1 illustrates five situations (V1–V5) that may occur in the real
waste management system. It simulates fictional situations, but similar
as in the Prague-Mělník case with few additional possibilities. There are
V. Nevrlý et al.
Fig. 1. Situations for facility and route decision.
three potential locations for the WtE plant (A, B, C). With regard to the
total cost, V1 is the most economic choice for waste disposal at the
producer's premises. It is because of the lowest gate-fee at location B
and no waste transport cost. Local garbage collection is assumed to be
identical among V1 through V5 and therefore it is not considered.
However, such an option does not include the emissions produced in
the highly-populated city due to the treatment process. V2 then re-
presents the second-best choice in terms of cost as a consequence of
shortest transport distance (25 km) and more affordable price (gate-fee:
80 EUR/t), but the waste is transported through another populated city.
Similarly, V3 involves waste transport via a smaller city but with a
higher price due to longer distance (50 km) even though the unit
transportation cost is lower. Transport of waste in V4 going to the plant
C bypasses the city with nearly no impact on the inhabitants and results
on the total cost of 88 EUR/t. Finally, V5 outlines the scenario where
the produced waste is treated in the WtE plant A at a reasonable cost of
87.75 EUR/t and without passing through any city. Furthermore, the
location A is not built in a populated area and the impact on the po-
pulation would thus be the lowest.
When using multi-objective approach, the aim is include also the
environment, i.e., the impact of additional emissions on the population.
Each of the scenarios is of different sustainability because the en-
vironmental point of view should be included in the decision-making
process with the reasonable cost increase. The accepted threshold
value, which is the number of impacted citizens, has a direct effect on
plant siting.
Decision-making based on the previously demonstrated approach
should be integrated into an advanced tool, which would consider local
air pollution, impacts on inhabitants, and cost. The application may
bring the following benefits for end users:
• reasonable trade-off between cheap and environmental solution;
• a solution with the lowest impact on citizens; and
• a precaution against the NIMBY (Not-In-My-BackYard) effect.
In summary, such an advanced tool would contribute to a better
system design with minimized impact on the citizens and would lead to
new research challenges. These comprise the relevancy of criteria,
model building, implementation, data availability, and input data pre-
processing. It is evident that such a model would be demanding in terms
of input data as well as the complexity and computational time. Table 1
summarises input parameters that are required for a proper solution. It
also shows to what extent these inputs are covered in this paper.
79
Journal of Environmental Management 220 (2018) 77–86
2.3. First step towards the advanced tool
The first and crucial step towards the development of the advanced
tool is an approach to handling point and line emission sources, dis-
persion of emissions, and the assessment of local impact on the popu-
lation. In this paper, an important step is provided through an integral
part of the model being a network composed of nodes interconnected
with edges. The above-mentioned input data are utilised to build an
edge characterisation procedure, which comprises production of pol-
lutants, road segmentation, dispersion, and concentration, to assess the
resulting impact. This provides valuable edge properties so that the
total edge impact can be compared and the best combination of edges
can be chosen in the final planning.
Two various sources, treatment facilities (point sources) and trans-
portation (line sources), are involved in the production of emissions in
the field of waste management. In the case of treatment facilities such
as WtE, the amount of produced emissions is based on plant capacity,
waste composition, facility design, and particularly on technologies
used for flue gas cleaning. These are facility-related emissions, often
measured on-site and even regulated by legislation, and represent re-
liable inputs to the calculation in terms of data quality (see Section
2.5.1). By contrast, emissions from the transport of waste are not easily
estimated or measured. The approach adopted in this paper therefore
relies on the estimates calculated from tabulated data and on mathe-
matical models.
In general, the production of emissions during treatment of waste is
several times larger than during its transport (see below), but building
the waste treatment facilities is necessary from the environmental point
of view. The preferred sites for these facilities are bound to utilise the
produced heat (industrial heat supply or district heating), while the
number of such places is limited. WtE also indirectly replaces fossil fuel
combustion and related emissions. Finally, in order to select a suitable
location from the perspective of the emission impact on the population,
both the WtE plant and the related waste transportation play important
roles.
An initial model of emission production from waste transportation is
described in section 2.5.2. For each route, the model takes into con-
sideration the route altitude profile, vehicle load, as well as the specific
type of vehicle used to determine the respective emission production.
Even though some assumptions are being made in the model, it is be-
lieved to be relevant regarding the objective of this paper. The emission
model is further enhanced to estimate the impact on the population in
the vicinity of the routes (emission source), which considers the road
segmentation and distances between a source and residential houses.
This approach will be demonstrated on the real route.
V. Nevrlý et al. Journal of Environmental Management 220 (2018) 77–86

Table 1
Input data required for advanced model.
Input data Addressed by this Comment
paper

Population data (location of inhabitants)
Exposure of the population to emissions
Road infrastructure (distances, route
profiles, etc.)
Model of emission production – Treatment
Dispersion model – Treatment
Model of emission production –
Transportation
Dispersion model – Transportation
Possible facility locations
Current air quality and emission limits
Cost (transport, treatment, etc.)
Waste quantities (current state and
prediction)
Legislation
Partially A lower level of detail, a residential area is described using one or more clusters. The same methodology
may be used with an increased level of detail. Computational time of pre-processing phase will increase
linearly.
Partially Exposure is based on the distance between the source of pollution and the population, and on the effects of
individual pollutants on the population.
Fully Based on real GIS data for the selected region (the Czech Republic).
Partially Complex modelling is needed to set up the model. Direct emissions from WtE and also the avoided
emissions due to fossil substitution should be considered.
No The calculation procedure is the same as in the transportation case. Dispersion studies are carried out using
a specialized software.
Partially Takes into account the route profile and vehicle load – based on the engine combustion equation and limits
for exhaust gases determined by the legislation.
Partially Simplified linear description of dispersion, homogenous uniform dispersion into all directions assumed.
Prevailing local wind direction is not considered. Complex dispersion modelling, for example via CFD, and
realistic values (slope, maximum) should be utilized, see (Pospisil and Jicha, 2011).
No The overall allocation model is not considered in this paper.
No Limits and short-listing of candidate locations due to currently high emission backgrounds are issues to be
handled in future work.
No Techno-economic models have already been provided by Gregor et al. (2017) for transportation, Ferdan
et al. (2015) for investment risks, and Ferdan et al. (2017) for emission trading.
No Could be estimated using the Justine tool presented in (Pavlas et al., 2017) or via methods from (Ghinea
et al., 2016)
No Could be included via taxes and bans of unwanted treatment methods.

2.4. Production of emissions from waste-related operation
This section describes the production of emissions from treatment
and transportation of waste. The contributions from both sources are
quantified and discussed in the context of local environmental impact.
2.4.1. Waste-to-energy plant
The present paper considers treatment of MMW and emissions
emitted from the WtE plant, see (Hwang et al., 2017) for greenhouse
gases produced in Korea. The amount of produced emissions is based on
internal parameters and technologies. Table 2 lists operating data from
an existing WtE plant in the Czech Republic, which produces both heat
and power, see (Pavlas et al., 2010).
Such facilities are commonly built within existing heat or power
plants. The combustion of fossil fuels is replaced by MMW-based en-
ergy, which might lead to positive effects (for example the overall GWP
impact might be negative), see (Ferdan et al., 2017) for a detailed ex-
planation. However, also other aspects should be incorporated into the
advanced allocation tool. As mentioned in Table 1, the facility is not
considered in this paper.
2.4.2. Road transportation
As mentioned by Ghiani et al. (2014), network representation of the
studied problem via a directed graph is the key element of the overall
model. An example of such a graph is shown in Fig. 2. Because each
road has its specific properties, evaluation of traffic capacity and
Table 2
Direct emissions from WtE plant (capacity 100 kt/y, operating time 8000 h/y).
CO NOx PM
WtE plant [kg/y] 22,272 87,840 2986
Specific emissions from MMW [kg/t] 0.223 0.878 0.030
Note: The WtE plant employs a backpressure turbine; however, comparable
data would be obtained if a condensing turbine were present (thermal and flue
gas cleaning subsystems are similar). The indirect impact of combined heat and
power production is energy saved in fossil-fuelled energy-producing plants, see
(Ferdan et al., 2017) for CO2 equivalent and (Tascione et al., 2016) for other
emissions characteristics.
passability is crucial to the identification of potential vehicle routes.
These properties include speed limits, maximum permitted loads,
height limitations, as well as other factors that determine the ability of
the vehicle to pass through. Such edges form the network.
In case of a bipartite graph, the connection between each producer
and treatment plant is represented by an edge. This edge is weighted
and its properties are defined by features of the road segments.
Considering only the economic pillar of sustainability, weight re-
presents the cost of using this edge, which is how it is included into the
objective function. To keep the task acceptable regarding computa-
tional time, the size of the graph (i.e., the number of nodes and edges) is
kept at a minimum, see Lam et al. (2011) for reduction techniques. The
real connection between two nodes consisting of several road segments
is aggregated and parameters of individual segments are transformed
into a single weight sets.
The goal of this section is to describe the main part of the emission
calculation which focuses on transportation in a specific area. As stated
by Gregor et al. (2017), the cost weight of an edge is dependent on the
amount of waste transported along the edge. In other words, this cost is
a function of the variable in the allocation problem. Similarly, traffic-
related emissions are expected as a function of load. The more waste is
transported, the larger amount of emissions is produced. In addition,
the respective fuel consumption varies according to the actual load,
vehicle type, slope, speed, etc., and thus the route altitude profile re-
presents another important input. The result in terms of emissions is the
assessment of the amount of pollutants (CO2, CO, NOx, HC + NOx, and
PM) produced per a specific edge consisting of multiple segments.
According to Table 3, emission limits for pollutants (except for CO2)
are set per engine emission classes. As for CO2, this needs to be calcu-
lated from fuel consumption. The production of other emissions was
SOx calculated from the limiting values in Table 3.
In general, emissions of pollutants and fuel consumption depend on
26,592
several factors that are very difficult to predict, and completely in-
0.266
dependent of each transport route. These factors are average speed per
the entire route, the manner of shifting speeds and its accuracy, engine
temperature and fluency of driving. For this paper, a simplified calcu-
lation was performed. It is based on the average theoretical consump-
tion, which is later on corrected according to the actual route profile
(inclination coefficient) and vehicle load (load coefficient).

80
V. Nevrlý et al. Journal of Environmental Management 220 (2018) 77–86

Fig. 2. Edge-related analysis.

Table 3 theoretical combustion of 28 L/100 km, the average theoretical pro-

Diesel emission limits according to legislation [European emission standards]. duction is 751 g/km for diesel and 655 g/km for petrol. The other
Emission class CO2 [g/ CO [g/ NOX [g/ HC + NOX [g/ PM [g/
pollutants (CO, NOx, HC + NOx, PM) follow the same pattern.
km] km] km] km] km] Subsequent parts of this paper are focused on emissions from waste
transport, their dispersion to the environment, and the impact on the
EURO I – 3.16 – 1.13 0.180 population.
EURO II – 1.00 – 0.70 0.080
EURO III – 0.64 0.50 0.56 0.050
EURO IV – 0.50 0.25 0.30 0.025 2.5. Dispersion and impact on the population
EURO V – 0.50 0.18 0.23 0.005
EURO VI – 0.50 0.08 0.17 0.005
In the previous section, two sources of pollutants, i.e., the WtE plant
and the transport vehicle, were discussed. The manner in which
harmful gases and PM are spread and dispersed in the environment is
The main goal of the emission model is to determine the production
quite an important topic, but it is very difficult to describe the process.
of pollutants based on the amount of waste and transport distance. The
Detailed mathematical models are usually based on measurements of
transport route is divided into smaller segments with dynamic length
many factors at different locations. These factors include wind speed
determination based on the local route profile (from 10 to 200 m) and
and direction, altitude profile of the terrain, distance from the emission
then the calculation is carried out for each of them. The inclination
source and its height, or other external influences. Depending on the
coefficient is determined according to the slope of the road segment
utilised mathematical approach, the obtained results can vary. In some
(see Fig. 3). The value 1 is valid for the theoretical consumption (28 L/
cases, e.g. when using a data mining approach (neural networks etc.),
100 km) for a lorry without any load, i.e., it corresponds to the theo-
the model is a black box in which the concentrations of substances in
retical value specified by the manufacturer.
the air are evaluated at selected locations based on the input data. In
For the load coefficient, it is necessary to prepare individual graphs
the ideal case the output can be written as a function. The results of
for each of the typical lorries. A theoretical experiment was done with a
modelling are diagrams describing the concentrations/intensities and
load of 24 t of waste (load coefficient was 1.6). The maximum load for
amounts of pollutants at specific locations.
garbage trucks is approx. 10 t and for lorries, with pressed waste, it is
The impact on the population is considered from the source of
approx. 25 t. The coefficient without load is equal to 1, while it linearly
pollution if it is within a particular area, which is determined according
increases up to the maximum load, where it is equal to 1.6.
to a predefined or calculated threshold (the furthest affected point). The
Then, the CO2 production was calculated according to the metho-
threshold is generally a function of direction in which the emissions
dology of the Environmental Protection Agency (EPA – www.epa.gov).
spread (wind), surrounding terrain (geographical structure), or other
Emissions of CO2 were predicted from the combustion equation for
variables. However, also the type of the pollutant plays an important
diesel or petrol engines.
role. Pollutants have different weights and their dispersion varies a lot.
The next step in the calculation procedure is to consider complete
This was dealt with in detail by Zavila et al. (2014), while Jícha et al.
combustion (i.e., 100% of the fuel is burnt) and the ratio of the molar
(2000) focused on dispersion in a street canyon and an adjacent urban
atomic weight of carbon and molecular weight of CO2. Given the
area.

Fig. 3. Main characteristic of angle inclination (Fischer, 2016).

81
V. Nevrlý et al. Journal of Environmental Management 220 (2018) 77–86

Fig. 4. Theoretical characterization of the affected area by functions.

Fig. 5. Illustration of segment impact calculation.

Fig. 6. Altitude profile of the Olomouc–Brno route.

The methodology of this paper is further introduced starting with segment. The furthest affected area (threshold distance – Tα ) can vary
Fig. 4a, which shows a theoretical area that is affected by emissions due to local conditions (wind direction, terrain profile etc.). Threshold
from a specific location denoted by the source point P. It also corre- distance Tα according to Fig. 4a is shown in the graph in Fig. 4b. Fig. 4c
sponds with the following Fig. 5, where it is applied for the specific then illustrates the Gaussian dependences of emission concentration on

82
V. Nevrlý et al.
Fig. 7. Area of impact of the pollutants for the 10 km threshold (map source: www.mapy.cz).
the distance from the source point P for α1 and α2 from Fig. 4b. Func-
tional dependency and impact of waste transportation should be com-
puted for each considered segment as shown in Fig. 5, where it is ne-
cessary to keep in mind the changing place of production during the
movement of the vehicle.
In the scope of the advanced location model, each transport route
(A–B) is divided into smaller segments (ab1, ab2, etc.) with lengths
adjusted according to the profile. The emission rate for the vehicle
hauling its load along each of the segments was evaluated according to
the methodology from the previous section. Each individual segment is
therefore considered to be a source point, for which the dispersion area
has to be described on the basis of Fig. 4a–c.
A further step is to propose an original methodology for segment
impact calculation. In general, each pollutant has its own threshold and
concentration function. In this paper, the methodology for segment
83
Journal of Environmental Management 220 (2018) 77–86
impact calculation was simplified from the realistic case by considering
a circular dispersion, compare Fig. 5a and b, and the concentration
which is given by one linear function Cd (independent of α ), unlike in
Fig. 4c, for all pollutants.
Next, the investigated geographical area along the edge is clustered
and the number of residents living in each of the clusters is evaluated.
The distances di, j between all clusters and all segments were calculated
with their GPS coordinates, while the Spherical Law of Cosines was
used. The distance then enters function Cd and, together with the
number of residents, the impact of individual pollutants is calculated.
The overall impact for each pollutant is then obtained by aggregating
the segments along the route A–B.
According to the previously mentioned calculations, the whole
procedure consists of the following steps:
V. Nevrlý et al.
Fig. 8. The impact of pollutants on population along the Olomouc–Brno route.
Fig. 9. The impact of pollutants on population along the Brno–Olomouc route.
1) Road division into segments and emission production calculation for
each segment.
2) Distance calculation between segment representatives and clusters
of residents.
3) Assignment of the concentration to specific municipality and mul-
tiplication by the number of residents.
4) Linear approximation and data aggregation for all the segments.
The result is one of the required inputs to the advanced location
model. However, it can be also used for other types of tasks such as
routing problems, where for example the paper by Bektaş and Laporte
(2011) might be extended.
3. Results/case study
One particular route was chosen for the purpose of presenting the
results. This is a real route via which MMW is transported ca. 800 times
a year from a transfer station located in the city of Olomouc to the WtE
plant in Brno (both located in the Czech Republic) with the distance
being ca. 81 km. The analysis concerns a vehicle – a lorry with a tri-axle
trailer – loaded with two 30 m3 (ca. 12.2 t) waste containers and
84
Journal of Environmental Management 220 (2018) 77–86
heading from the transfer station in Olomouc to the WtE plant in Brno,
which subsequently returns empty back to Olomouc. It is assumed that
this vehicle falls into the EURO V class. Different load coefficients are
considered for driving in both directions. Route profiling is shown in
Fig. 6, where the maximum altitude is ca. 300 m and the lowest is ca.
210 m.
Coefficients according to Fig. 3 for diesel consumption corrections
were applied. The estimated diesel consumption for the trip from
Olomouc to Brno was approx. 48 L, which was later confirmed by the
operator. Consumption on the route back (i.e. without load) was found
to be approx. 30 L (i.e., 35 L/100 km).
For the emission dispersion calculation, the concentration function
was determined to be linear in contrast to Fig. 4c. The maximum dis-
tance (threshold), up to which pollutants have any effect, was set to
10 km according to Shiva Nagendra et al. (2016). Municipalities along
the route were considered as clusters. The locations of the munici-
palities along this route are shown in Fig. 7. The hatched area indicates
the affected territory and municipalities. It means that the whole area
surrounded by the red line is impacted by the transportation along the
specified route with the produced pollutants.
The impact of pollutants produced by the discussed vehicle on the
V. Nevrlý et al. Journal of Environmental Management 220 (2018) 77–86

Fig. 10. Impact of PM on population – both ways.

Table 4 vicinity of a source of pollution is affected. This paper presents an initial

The impacts of production of emissions by the EURO V vehicle per route and step for calculating the emission levels in reverse logistics problems in
population. waste management to support decision-making based on the impact of
Pollutant 103 CO2 CO NOX HC + NOX PM pollutants. The analysis of waste transportation is also necessary in
terms of designing a complex approach. The planning process requires
Olomouc-Brno [kg∙cap] 5183 3450 1587 1242 34 reverse logistics-based decisions. Network flow models and facility lo-
Brno-Olomouc [kg∙cap] 3286 2187 1006 787 22
cation problems are usually used in such a situation. This paper pro-
Specific emissions per MMW 0.0155 0.0103 0.0047 0.0037 0.0001
for the entire route [kg/ poses a study which is crucial for the preparation of the necessary input
t] data regarding the underlying network edge properties.
The calculation of distance, load and route altitude profile-based
emission data was proposed for segmented route. The segments were
population is shown in Fig. 8 through Fig. 10. Figs. 8 and 9 depict the used for final impact calculation using dispersion, distances from
impact of pollutants CO2, CO, HC-NOX and NOX on population along source, number of residents and their location, and threshold. The basic
the Olomouc–Brno (Brno-Olomouc) route. The individual pollutants are impact of pollutants from MMW transportation on the population was
distinguished by colours, while the smallest impact in terms of quantity obtained and quantified for the specific route (Olomouc–Brno). The
is caused by NOX and the largest by CO2. However, taking into account analysed pollutants were CO2, CO, NOx, HC + NOx, and PM with the
the toxicity for the inhabitants, a further recalculation is needed to actual impacts for the entire route being 8469 ∙ 103 kg ∙ cap, 5637 kg ∙
unify the impact of pollutants. The impact of PM is shown in Fig. 10 for cap, 2593 kg ∙ cap, 2029 kg ∙ cap, and 56 kg ∙ cap, respectively. The basic
scale and clarity reasons for both directions. From the graphs it is clear concept of a model for which these results apply has also been in-
that the highest impact is near the route terminal cities (Brno, Olo- troduced with specific requirements of combining both global and local
mouc), which are highly populated areas. However, the calculation also aspects.
takes into account the distribution of other municipalities along the The presented models can be extended in the future in a variety of
route. The influences of vehicle load and road inclination are also im- ways. More accurate dispersion function (for concentration calculation)
portant factors. reflecting the prevailing wind direction as well as a threshold based on
Table 4 summarises the overall effects of each pollutant in both real measurements and more detailed analysis might be utilised. The
route directions. calculation of the impact of pollutants can make use of the detailed
For the characterization of the route in terms of all pollutants, we location data regarding residential houses. Other factors could be in-
can draw inspiration from the LCA methodology. For instance, GWP cluded in this impact as well.
may be used to describe the effects of greenhouse gas and other emis- There are other research steps that should also be taken. Dispersion
sions for assessment of the impact on humans. The latter index would studies should be carried out for all WtE plant candidate localities in
then be calculated from the results listed in Table 4 and then it would terms of various plant capacities. The calculation should consider
serve as one of the input parameters of the mathematical model. This avoided emissions (due to the replacement of fossil fuels) in addition to
calculation is required for all the routes (edges) that form the network. the actual ones. The impact on the population should be calculated for
In case of Czech Republic with vertices representing centres of micro- all the underlying network edges. Last but not least, the mathematical
regions, the network comprises 206 vertices and ca. 2000 edges, see model should be implemented using an optimisation tool such as
(Ferdan et al., 2015) for a detailed description. GAMS.

4. Conclusion Acknowledgement

The evaluation of environmental impacts is required when dealing This research has been supported by EU project Sustainable Process
with waste management systems. The majority of recent studies utilised Integration Laboratory – SPIL funded as project No. CZ.02.1.01/0.0/
a global measure, i.e. GWP, as the impact metric. However, a local- 0.0/15_ 003/0000456, by Czech Republic Operational Programme
based analysis plays an important role because the population in the Research, Development and Education, Priority 1: Strengthening

85
V. Nevrlý et al.
capacity for quality research.
References
Bektaş, T., Laporte, G., 2011. The pollution-routing problem. Transp. Res. Part B
Methodol. 45 (8), 1232–1250. http://dx.doi.org/10.1016/j.trb.2011.02.004.
Bernstad, A., Jansen, J.C., 2012. Review of comparative LCAs of food waste management
systems – current status and potential improvements. Waste Manag. 32 (12),
2439–2455. http://dx.doi.org/10.1016/j.wasman.2012.07.023.
Bing, X., Bloemhof, J.M., Ramos, T.R.P., Barbosa-Povoa, A.P., Wong, C.Y., van der Vorst,
J.G.A.J., 2016. Research challenges in municipal solid waste logistics management.
Waste Manag. 48, 584–592. http://dx.doi.org/10.1016/j.wasman.2015.11.025.
Cai, B., Li, W., Dhakal, S., Wang, J., 2018. Source data supported high resolution carbon
emissions inventory for urban areas of the Beijing-Tianjin-Hebei region: spatial pat-
terns, decomposition and policy implications. J. Environ. Manag. 206, 786–799.
http://dx.doi.org/10.1016/j.jenvman.2017.11.038.
Cheng, Z., Li, L., Liu, J., 2017. The emissions reduction effect and technical progress effect
of environmental regulation policy tools. J. Clean. Prod. 149, 191–205. http://dx.doi.
org/10.1016/j.jclepro.2017.02.105.
Cleary, J., 2009. Life cycle assessments of municipal solid waste management systems: a
comparative analysis of selected peer-reviewed literature. Environ. Int. 35 (8),
1256–1266. http://dx.doi.org/10.1016/j.envint.2009.07.009.
Čuček, L., Klemeš, J.J., Kravanja, Z., 2012. A review of footprint analysis tools for
monitoring impacts on sustainability. J. Clean. Prod. 34, 9–20. http://dx.doi.org/10.
1016/j.jclepro.2012.02.036.
Danny Harvey, L.D., 1993. A guide to global warming potentials (GWPs). Energy Policy
21 (1), 24–34. http://dx.doi.org/10.1016/0301-4215(93)90205-T.
Emission standard EURO. Autolexicon, www.autolexicon.net/cs/articles/emisni-norma-
euro/(accessed 2017-06-14).
Ferdan, T., Šomplák, R., Zavíralová, L., Pavlas, M., Frýba, L., 2015. A waste-to-energy
project: a complex approach towards the assessment of investment risks. Appl.
Therm. Eng. 89 (1), 1127–1136. http://dx.doi.org/10.1016/j.applthermaleng.2015.
04.005.
Ferdan, T., Pavlas, M., Šomplák, R., Nevrlý, V., Klemeš, J.J., Pitschke, T., 2017. Allocation
analysis of waste treatment nodes for economic optimisation considering reduced
greenhouse gas emissions. Chem. Eng. Trans. 61. http://dx.doi.org/10.3303/
CET1761159.
Fischer, W., 2016. Possibilities and insights in municipal area and tour planning. In: In:
Pomberger, Roland (Ed.), Montanuniversität Leoben (Hg.): Tagungsband zur, vol. 13
Recy & DepoTech-Konferenz 391–196, 2016, (in German).
Ghiani, G., Laganà, D., Manni, E., Musmanno, R., Vigo, D., 2014. Operations research in
solid waste management: a survey of strategic and tactical issues. Comput. Oper. Res.
44, 22–32. http://dx.doi.org/10.1016/j.cor.2013.10.006.
Ghinea, C., Drăgoi, E.N., Comăniţă, E.-D., Gavrilescu, M., Câmpean, T., Curteanu, S.,
Gavrilescu, M., 2016. Forecasting municipal solid waste generation using prognostic
tools and regression analysis. J. Environ. Manag. 182, 80–93. http://dx.doi.org/10.
1016/j.jenvman.2016.07.026.
Goulart Coelho, L.M., Lange, L.C., Coelho, H.M.G., 2017. Multi-criteria decision making
to support waste management: a critical review of current practices and methods.
Waste Manag. Res. 35 (1), 3–28. http://dx.doi.org/10.1177/0734242X16664024.
Gregor, J., Pavlas, M., Šomplák, R., 2017. Transportation cost as an integral part of supply
chain optimization in the field of waste management. Chem. Eng. Trans. 1927–1932.
http://dx.doi.org/10.3303/CET1756322.
Havukainen, J., Zhan, M., Dong, J., Liikanen, M., Deviatkin, I., Li, X., Horttanainen, M.,
2017. Environmental impact assessment of municipal solid waste management in-
corporating mechanical treatment of waste and incineration in Hangzhou, China. J.
Clean. Prod. 141, 453–461. http://dx.doi.org/10.1016/j.jclepro.2016.09.146.
He, R., Xiong, Y., Linc, Z., 2016. Carbon emissions in a dual channel closed loop supply
chain: the impact of consumer free riding behavior. J. Clean. Prod. 134 (Part A),
384–394. http://dx.doi.org/10.1016/j.jclepro.2016.02.142.
86
Journal of Environmental Management 220 (2018) 77–86
Hombach, L.E., Cambero, C., Sowlati, T., Walther, G., 2016. Optimal design of supply
chains for second generation biofuels incorporating European biofuel regulations. J.
Clean. Prod. 133, 565–575. http://dx.doi.org/10.1016/j.jclepro.2016.05.107.
Hwang, K.-L., Choi, S.-M., Kim, M.-K., Heo, J.-B., Zoh, K.-D., 2017. Emission of green-
house gases from waste incineration in Korea. J. Environ. Manag. 196, 710–718.
http://dx.doi.org/10.1016/j.jenvman.2017.03.071.
Jícha, M., Pospíšil, J., Katolický, J., 2000. Dispersion of pollutants in street canyon under
traffic induced flow and turbulence. Environ. Monit. Assess. 65, 343–351. http://dx.
doi.org/10.1023/A:1006452422885.
Kanzian, C., Kühmaier, M., Zazgornik, J., Stampfer, K., 2013. Design of forest energy
supply networks using multi-objective optimization. Biomass Bioenergy 58, 294–302.
http://dx.doi.org/10.1016/j.biombioe.2013.10.009.
Lam, H.L., Klemeš, J.J., Kravanja, Z., 2011. Model-size reduction techniques for large-
scale biomass production and supply networks. Energy 36 (8), 4599–4608. http://dx.
doi.org/10.1016/j.energy.2011.03.036.
MacLean, H.L., Lave, L.B., 2003. Evaluating automobile fuel/propulsion system tech-
nologies. Prog. Energy Combust. Sci. 29 (1), 1–69. http://dx.doi.org/10.1016/S0360-
1285(02)00032-1.
Olsthoorn, X., Tyteca, D., Wehrmeyer, W., Wagner, M., 2001. Environmental indicators
for business: a review of the literature and standardisation methods. J. Clean. Prod. 9
(5), 453–463. http://dx.doi.org/10.1016/S0959-6526(01)00005-1.
Pavlas, M., Touš, M., Bébar, L., Stehlík, P., 2010. Waste to energy – an evaluation of the
environmental impact. Appl. Therm. Eng. 30 (16), 2326–2332. http://dx.doi.org/10.
1016/j.applthermaleng.2009.10.019.
Pavlas, M., Šomplák, R., Smejkalová, V., Nevrlý, V., Zavíralová, L., Kůdela, J., Popela, P.,
2017. Spatially distributed production data for supply chain models - forecasting with
hazardous waste. J. Clean. Prod. 161, 1317–1328. http://dx.doi.org/10.1016/j.
jclepro.2017.06.107.
Pospisil, J., Jicha, M., 2011. Particle re-suspension in street canyon with two-way traffic.
Int. J. Environ. Pollut. 44 (1–4), 271–279. http://dx.doi.org/10.1504/IJEP.2011.
038427.
Samanlioglu, F., 2013. A multi-objective mathematical model for the industrial hazardous
waste location-routing problem. Eur. J. Oper. Res. 226 (2), 332–340. http://dx.doi.
org/10.1016/j.ejor.2012.11.019.
Shiva Nagendra, S.M., Diya, M., Chithra, V.S., Menon, J.S., Peter, A.E., 2016.
Characteristics of air pollutants at near and far field regions of a national highway
located at an industrial complex. Transp. Res. Part D Transp. Environ. 48, 1–13.
http://dx.doi.org/10.1016/j.trd.2016.07.006.
Šomplák, R., Pavlas, M., Kropáč, J., Putna, O., Procházka, V., 2014. Logistic model-based
tool for policy-making towards sustainable waste management. Clean Technol.
Environ. Policy 16, 1275–1286. http://dx.doi.org/10.1007/s10098-014-0744-5.
Tascione, V., Mosca, R., Raggi, A., 2016. Optimizing the environmental performance of
integrated waste management scenarios by means of linear programming: a case
study. J. Clean. Prod. 112 (Part 4), 3086–3096. http://dx.doi.org/10.1016/j.jclepro.
2015.10.016.
Wirl, F., 2012. Global warming: prices versus quantities from a strategic point of view. J.
Environ. Econ. Manag. 64 (2), 217–229. http://dx.doi.org/10.1016/j.jeem.2011.11.
002.
Yilmaz, O., Kara, B.Y., Yetis, U., 2017. Hazardous waste management system design
under population and environmental impact considerations. J. Environ. Manag. 203
(Part 2), 720–731. http://dx.doi.org/10.1016/j.jenvman.2016.06.015.
Zavila, O., Bojko, M., Bitta, J., Pavlíková, I., Jančík, P., Dobeš, P., 2014. Influence of
complex terrain and meteorological conditions on gas pollutant dispersion in large
domains - comparative mathematical study. Spektrum 14, 1.
Zhang, X., Huang, G., 2014. Municipal solid waste management planning considering
greenhouse gas emission trading under fuzzy environment. J. Environ. Manag. 135,
11–18. http://dx.doi.org/10.1016/j.jenvman.2014.01.014.
Zohal, M., Soleimani, H., 2016. Developing an ant colony approach for Green closed-loop
supply chain network design: a case study in gold industry. J. Clean. Prod. 133,
314–337. http://dx.doi.org/10.1016/j.jclepro.2016.05.091.