Calculating the costs of waste collection: A methodological proposal

Abstract

Waste collection and transport can generate up to 70% of the total costs of the system. Separated collection of recyclables implies additional costs for which the sale of recycled waste often does not compensate, but there is increased pressure to reach the long-term recycling objectives set by law. The proper estimation and monitoring of waste collection costs are essential to define the most cost-effective waste collection system. The aim of this study is to propose and implement a management tool to determine waste collection costs for different waste collection schemes. Based on input data, such as waste quantity and composition, the number of waste bins, the location of collection points, the type of collection vehicle, crew, collection route, etc., the developed tool can calculate the time and costs of waste collection (per vehicle, collection point or tonne of collected waste). This tool uses Excel spreadsheets and it was tested on a district in the central area of the city of Kragujevac to calculate the costs of waste collection for two scenarios: Collecting all waste as mixed waste, and collecting separately recyclables and residual waste. The developed tool can be useful for municipal solid waste management companies, since it allows benchmarking and variance analysis.

Keywords

Waste collection costs of waste collection separate collection cost effectiveness energy efficiency

Introduction

Municipal solid waste (MSW) management is a multidisciplinary activity that includes generation, storage and collection, transport, treatment and waste disposal. Waste collection and transport can generate up to 70% of the MSW system costs (Greco et al., 2015; Tavares et al., 2009). The proper estimation and monitoring of the waste collection costs are essential to define the most cost-effective waste collection system (Huang et al., 2011; Jacobsen et al., 2012). This can also increase the waste collection efficiency and reduce tax rates paid by citizens.

Numerous studies can be found that analyse the costs of MSW management in different countries that propose methods and tools for measuring financial performance of waste management (Pires et al., 2011). D’Onza et al. (2016) analyse full cost accounting (FCA) in 68 Italian municipalities. They developed a FCA management tool to calculate full collection costs of separated waste collection. Greco et al. (2015) show that the economies of scale and cost drivers differ across the types of waste. Separated waste collection implies additional costs, for which the sale of recycled waste often does not compensate (D’Onza et al., 2016). But money is not the only means of motivating people to recycle. Many municipalities were forced to assess their solid waste management programmes in order to reach recycling legislative targets. Larsen et al. (2010) conducted research in order to discover how much the recycling rate can be increased through improvements of collection schemes when organisational and technical limitations are respected, as well as, what the environmental and economic consequences will be. According to them, costs of collection and treatment are reduced by increasing recycling, because the high costs for incineration are avoided.

Waste management in the Republic of Serbia, as well as in other developing countries, is still based on landfilling. Waste management practice can be characterised as undeveloped, as waste management consists of waste collection and land disposal only (Stanisavljevic et al., 2012). According to the latest report from Serbian Environmental Protection Agency (SEPA) (2014), 20% of the population is still not covered with organised waste collection, so the main challenge is to increase coverage of waste collection, as well as to transfer from waste dumping into sanitary landfilling. In previous years, several sanitary landfills were opened, and the task that Serbia is currently facing, developed countries have dealt with in the past. In 2009, Serbia applied to become a member state of European Union (EU), so it must fulfil all requirements according to EU policy. The great challenge is to harmonise national legislation with EU directives. Based on the EU Landfill Directive (Council Directive 1999/31/EC), a new Serbian waste landfill regulation (Official Gazette of RS 92/10) was adopted. According to this, the share of biodegradable waste that can be landfilled is to be decreased by 25%, 50% and 65% until the years 2016, 2019 and 2026, respectively, so this type of waste has to be collected separately. The recycling target set for 2014 is 25% of the packaging waste put on the market, and according to SEPA this target was achieved (Serbian Environmental Protection Agency, 2014). The share of separately collected recyclables must also increase in the following years. These also affect the price of waste collection and treatment.

When talking about the waste charges that citizens are called upon to pay, it must be mentioned that in many municipalities the price is lower than the economical price. The key lies in the fact that these companies have the support of a city’s budget through public subsidies (Jovicic et al., 2015). In previous years, there was a strong government call for privatisation of public services, looking at efficiency as the main reason, along with cutting public subsidies for covering losses. Hence, subsidies will definitely get lower in future years. On the basis of all the mentioned points, companies will be forced to assess their solid waste management programmes and raise the charges that citizens pay for collecting, transporting and processing solid waste.

In order to provide their services in the best possible way, proper cost accounting and management has become a critical issue for every MSW company. The use of different MSW schemes increases the complexity of waste management operations and complicates evaluation of the costs (Karagiannidis et al., 2008). Hence, the aim of the present study is to propose and implement a management tool to determine costs of waste collection. This tool should allow benchmarking and variance analysis, and it should be used as a fast method to determine time and costs of waste collection.

Materials and methods

Methodology

Costs that appear in the complete process of waste management could be categorised into seven main categories: Up-front, operating, back-end, remediation, contingent, environmental and social (United States Environmental Protection Agency, 1997). The entire lifecycle of MSW activities from the ‘cradle’ to the ‘grave’ are covered with the first three categories. Up-front costs include initial investments in necessary equipment for waste collection and transport. Operating costs are expenses of managing MSW on a daily basis, while back-end costs are expenses in proper care of landfills at the end of their useful lives. Since our analysis focuses on the waste collection process, we took into account only the costs that this process includes. According to previously mentioned classification, these costs include up-front costs (initial investments for purchasing the necessary equipment for waste collection – bins, vehicles and other types of equipment) and operating costs (costs of workforce, fuel and managing waste collection on a daily basis). Maintenance costs and depreciation rate for equipment is also included.

The aim of our study is to define a mathematical model and to implement it in a management tool that can provide a quick analysis of waste collection costs. To define a model we started from the typical scheme of waste collecting shown in Figure 1. The basic activities in the MSW collection and transport process could be divided into four different activity stages (Boskovic et al., 2013).

At the beginning of the work day, a vehicle starts from a garage and travels to a collection area. This is the first stage in the MSW collection and transport process.

The second stage starts by entering a collection area. A vehicle stops at the first collection point (CP), unloads the waste bins and then continues to another CP along a predetermined route. The length of a collection route is determined by the number of CPs that a vehicle can service before it is filled to capacity. After the vehicle is fully loaded, the second stage is finished.

The fully loaded vehicle drives to a treatment or disposal facility, which presents the third stage of a collection process. After a vehicle is unloaded, it drives back to the first CP of another collection route. Each collection route starts in the same way and ends by unloading the collected waste. At the end of a working day, a vehicle drives back to a garage, which represents the fourth stage of the process of MSW collection and transport.

The mathematical model presented in this article is based on the mathematical model for waste collection costs developed by Barlaz et al. (1995) and Curtis and Dumas (2000), and it is expanded and adapted to local conditions of waste collection in the Republic of Serbia. The developed management tool, for given input data calculates:

cost of waste collection per year;

cost of waste collection per collection vehicle per year;

cost of waste collection per CP per year;

cost of waste collection per tonne of collected waste.

Cost of waste collection per year (a) calculates the number of vehicles and waste bins used for waste collection, multiplied with annual vehicle and bin costs. This unit cost can be calculated for one part of the city or the entire city; it is in the function of the number of residential and commercial objects served and it will be described in detail in the following text.

Unit costs (b), (c) and (d) are independent of the number of served residential and commercial objects. Among these unit costs, a cost of waste collection per tonne of collected waste seems to be the most important for comparison of different waste collection schemes.

The parameter from which the cost calculation starts is the amount of generated waste for each CP. Since the value for frequency of waste collection is the number of collections per week, the amount of waste for each CP should be expressed in units per week. According to our previous research (Boskovic and Jovicic, 2015), the total amount of waste generated per week per CP can be calculated by using:

W_{t o t} [k g] = 7 \cdot (W_{r e s} + W_{c o m m})

(1)

where $W_{r e s} [k g]$ is the daily waste generated in residential activities and $W_{c o m m} [k g]$ is the daily waste generated in commercial activities.

Since the waste generated in residential activities depends on the number of citizens that live in the area ( $p [-]$ ) and daily waste generation per capita ( $w_{c a p} [k g]$ ), it follows that:

W_{r e s} = p \cdot w_{c a p}

(2)

The amount of waste generated in commercial activities depends of the type of activity and the surface that this activity covers. Hence, it can be calculated using:

W_{c o m m} [k g] = \sum_{i = 1}^{n} c_{w i} \cdot A_{i}

(3)

where n [−] is the number of different commercial activity classes, $c_{w i} [k g m^{- 2}]$ is the commercial activity coefficient and $A_{i} [m^{2}]$ is the total surface covered by commercial activity class.

Geographic information system (GIS) plays an important role in determining the quantity of waste that should be collected at each CP. By using the possibilities of GIS analysis, each building in the area under study is joined to the nearest CP. For the service area of each CP, quantity of generated waste can be calculated according to previously presented equations. Since, this research is a continuation of our previous research on developing the methodology for optimal CP locations and number of waste bins (Boskovic and Jovicic, 2015), detailed information about using of GIS in determining the quantity of generated waste, as well as determining the commercial activity classes, can be found in our references.

The methodology used in this research to calculate costs of waste collection starts with the calculation of the number of vehicles needed to service all CPs in the area under study and the time used for the collection process. Since the methodology could be used for determining the cost of collection of different waste collection schemes, in the following text the collection of mixed (undifferentiated) waste will be explained first and later expanded to separate collection.

If all waste is collected as mixed waste, then it follows that:

W_{m i x} = W_{t o t}

(5)

where $W_{m i x} [k g]$ is the weekly waste generation rate per CP. According to Barlaz et al. (1995), on whose method our methodology is based, the number of CPs that a vehicle can service before it is filled to capacity represents the number of CPs $N_{C P} [-]$ along one collection route and it can be calculated using:

N_{C P} = \frac{U_{c} \cdot V_{c} \cdot F_{c}}{W_{m i x} / ρ_{m i x}}

(6)

where $N_{C P} [-]$ is the number of served CPs, $U_{c} [-]$ is the vehicle utilisation factor (usable per total vehicle capacity), $V_{c} [m^{3}]$ is the vehicle capacity, $F_{c} [-]$ is the frequency of waste collection per week and $ρ_{m i x} [k g m^{- 3}]$ is the overall density of mixed waste.

Overall density of mixed waste can be calculated by:

ρ_{m i x} = \frac{1}{\sum_{i} \frac{ρ_{i}}{ρ_{c i}}}

(7)

where ρ_i is the weight fraction for waste component i and ρ_ci is the compacted density of waste component i.

The number of waste bins (N_wb) that a vehicle can service during one collection trip is defined as:

N_{w b} = \frac{N_{C P}}{a_{w b}}

(8)

where $a_{w b} [-]$ is the average number of waste bins per CP.

The number of serviced CPs (N_CP), multiplied by the amount of time that a vehicle spends to unload waste bins at each CP and travelling between locations, present the time that the vehicle spends in Stage II of waste collection and transport process, in accordance with Figure 1.

The length of time that a collection vehicle takes for one collection trip (T_C [min]), beside time taken for Stage II, includes time taken for Stages I and III, and time for unloading the vehicle at the disposal facility. It can be calculated by using:

T_{C} = T_{I} + T_{I I l} \cdot N_{C P} + T_{I I d} \cdot (N_{C P} - 1) + T_{I I I} + T_{u l}

(9)

where $T_{I} [m i n]$ is the length of time taken for Stage I, $T_{I I l} [m i n]$ is the loading time at a the CP, $T_{I I d} [m i n]$ is the time taken for driving between two CPs, $T_{I I I} [m i n]$ is the length of time taken for Stage III and $T_{u l} [m i n]$ is the time to unload the collection vehicle at the disposal facility.

The length of time that a collection vehicle takes to make a complete collection trip varies depending on the characteristics of the service area, distance from service area to unloading facilities, road and traffic characteristics, etc. Therefore, many inputs that flow into calculations must be determined for each case study, usually with support of field work and a geographic informational system.

The time taken for each stage of the waste collection and transport process ( $T_{i} [m i n]$ ) can be determined with distance travelled for a specific stage ( $D_{i} [k m]$ ), divided by average vehicle speed for that stage ( $V_{i} [k m h^{- 1}]$ ):

T_{i} = \frac{D_{i}}{V_{i}} \cdot 60

(10)

In order to define number of collection trips ( $N_{C T} [-]$ ) that a vehicle can make during one work shift, the time for lunch period, other breaks and time taken for Stage IV must be taken into account. Hence, it can be calculated as:

N_{C T} = \frac{N_{W H} / 60 - (T_{L} + T_{B} + T_{I V})}{T_{C}}

(11)

where $N_{W H} [h]$ is the number of work hours per shift, $T_{L} [m i n]$ is the lunch period, $T_{B} [m i n]$ is the break period and $T_{I V} [m i n]$ is the travel time in Stage IV.

The number of collection vehicle trips needed ( $N_{C T A} [-]$ ) to service all of the CPs is defined by:

N_{C T A} = \frac{N_{C P A}}{N_{C P}}

(12)

where $N_{C P A} [-]$ is the number of all CPs in the service area.

The next step is to define the number of collection vehicles ( $N_{V} [-]$ ) needed to service all CPs during one collection cycle:

N_{V} = \frac{N_{C T A}}{N_{C T}} \cdot \frac{F_{c}}{N_{W D}}

(13)

where $N_{W D} [-]$ is the number of workdays per week

Defining the costs of waste collection

As previously mentioned. the costs present a sum of capital and operating costs. Annual capital costs per collection vehicle can be calculated in accordance with:

C_{c v} = (1 + e) \cdot C_{v} \cdot C R F

(14)

where $C_{c v} [€ v e h i c l e^{- 1} y e a r^{- 1}]$ is the capital cost of vehicle amortised over its economic life, e[−] is the administrative rate, $C_{v} [€]$ is the unit price of vehicle and CRF is the capital recovery factor.

The capital recovery factor can be calculated as:

C R F = \frac{i \cdot {(1 + i)}^{L}}{{(1 + i)}^{L} - 1}

(15)

where i is the yearly discount rate and L [year] is the economic life of collection vehicle.

The annual capital cost per waste bin ( $C_{c w b} [€ y e a r^{- 1}]$ ) is defined as:

C_{c w b} = (1 + e) \cdot C_{v w b} \cdot C R F_{w b}

(16)

where $C_{v w b} [€ b i n^{- 1} y e a r^{- 1}]$ is the capital cost of waste bin amortised over its economic life, $C_{w b} [€]$ is the unit price of waste bin and CRF_wb is the capital recovery factor:

C R F_{w b} = \frac{i \cdot {(1 + i)}^{L w b}}{{(1 + i)}^{L w b} - 1}

(17)

where $L_{w b} [y e a r]$ is the economic life of waste bin.

The operating cost of the collection vehicle per year (C_ov $[€ v e h i c l e^{- 1} y e a r^{- 1}]$ ) is defined as:

C_{o v} = (1 + e) \cdot {(1 + a) [(1 + b_{w}) \cdot (S_{w} \cdot N_{w} + S_{d}) \cdot W_{h} \cdot N_{W D} \cdot \frac{365}{7}] + C + [d \cdot (N w + 1]}

(18)

where $C_{o v} [€ v e h i c l e^{- 1} y e a r^{- 1}]$ is the annual operating cost of vehicle, a is the fringe benefit rates, b_w is the backup rate for collection workers, $S_{w} [€ h^{- 1}]$ is the salary rate per hour per worker, $N_{w} [-]$ is the number of collection workers per vehicle, $S_{d} [€ h^{- 1}]$ is the driver salary rate per hour, $W_{h} [-]$ is the working hours per day, $C [€ v e h i c l e^{- 1} y e a r^{- 1}]$ is the annual vehicle operation and maintenance costs and $d [€ w o r k e r^{- 1} y e a r^{- 1}]$ is all other expenses.

The collection cost per year per vehicle is calculated as:

C_{v y} = [(1 + b_{v}) \cdot C_{c v}] + N_{w b} \cdot C_{c w b} + C_{o v}

(19)

where $C_{v y} [€ y e a r^{- 1}]$ is the annual cost of collection per vehicle and $b_{v} [-]$ is the backup rate for the collection vehicle (backup vehicle per collection vehicle).

The cost of waste collection per year ( $C_{t a} [€ y e a r^{- 1}]$ ) is defined as:

C_{t a} = N_{C T A} \cdot C_{v y}

(20)

The cost of waste collection per CP per year is calculated according to:

C_{C P} = \frac{C_{v y}}{N_{C P}}

(21)

The cost of waste collection per tonne of collected waste ( $C_{t} [€]$ ) is defined as:

C_{t} = \frac{C_{C P}}{W_{t o t}}

(22)

The cost of separate waste collection could also be calculated according to formulas given for mixed collection. In that case two calculations have to be made. The first step is to define the cost of the collection of waste that is collected separately, then to calculate the cost of the collection of residual waste. The amount of recyclable waste that has to be collected depends on waste composition and the capture rate of recyclables, for each waste component separately collected. The separation rate of recyclables can be expressed as:

f_{i r} = \sum_{i} c_{r i} \cdot f_{i}

(23)

where $f_{i r} [-]$ is the separation rate of recyclables, $c_{r i} [-]$ is the capture rate for waste component i and $f_{i} = W_{i} / W_{t o t}$ is the weight fraction for recyclable waste component.

The amount of total generated waste ( $W_{t o t}$ ) can be presented as a sum of recyclable waste ( $W_{r e c}$ ) and the rest of mixed waste ( $W_{m i x r}$ ):

W_{t o t} = W_{r e c} + W_{m i x r} = \sum W_{i} (R) + \sum W_{i} (M)

(24)

where $W_{i} (R) [k g]$ is the amount of recyclable component I and $W_{i} (M) [k g]$ is the amount of the rest of the mixed waste.

Density of recyclable waste ( $ρ_{r e c} [k g m^{- 3}])$ can be calculated according to:

ρ_{r e c} = \frac{\sum f_{i} \cdot c_{r i}}{\sum \frac{f_{i}}{ρ_{c i}} c_{r i}}

(25)

Density of residues waste presents:

ρ_{m i x r} = \frac{\sum f_{i} \cdot (1 - c_{r i})}{\sum \frac{f_{i}}{ρ_{c i}} (1 - c_{r i})}

(26)

After the amount of recyclables and residues and their density are calculated, the whole procedure for determining the time and cost for waste collection can be done according to the given mathematical model for mixed waste collection.

Results and discussion

According to the mathematical model given in the previous section, the management tool using an Excel spreadsheet was developed. This tool provides a calculation of waste collection time and cost based on input data. These data must be collected for the area under study by gathering information from the utility company and field work using Global Positioning System (GPS) equipment and a geographical informational system.

The developed tool was tested on a district in the city of Kragujevac. Waste management in the city Kragujevac is in the charge of utility company Cistoca and it is based on landfilling, except for the low rate primary separation of recyclables (Polyethylene terephthalate (PET), cans, glass). The average amount of mixed waste collected per year is about 50,000 t, and the average amount of recyclables waste collected per year is about 500 t, which confirmed the previously mentioned low separation rate. The reasons for this are a low citizen participation rate in recycling and a lack of waste bins for separate collection. All the collected waste is directly transferred to landfill in Jovanovac, which is 3 km away from the city centre. The city is empirically divided into 11 sectors (collecting zones). The area being examined in this article is shown in Figure 2. This part of the city belongs to Sector No. 1, with waste collection at a daily level. This is a very representative area, because it consists of a large number of residential buildings, offices, shops and restaurants, as well as a faculty and student dormitory. The same area was studied in our previous research, aiming to optimise the number of CPs and waste bins (Boskovic and Jovicic, 2015). After the optimisation was made, the number of waste bins of 1.1 m³ needed to service the whole area was set at 175, located at 88 CPs. This will be the input data in our calculation. The costs and time for waste collection is determined for two different scenarios.

Scenario 1: Collection of all waste as mixed waste without the separate collection of recyclables.

Scenario 2: Separate collection of recyclable and mixed waste.

The calculations of time and costs for mixed waste collection (Scenario 1), based on previously presented mathematical model, are shown in Table 1. As it can be seen from the table, it is divided into three parts: waste parameters, collection parameters and costs of waste collection. The values of input data are shown in regular font, while the values for parameters that are the results of calculations are shown in italic.

Table 1.

Calculation of waste collection costs of mixed waste.

Symbol	Description	Value
Waste parameters
w_cap[kg·capita⁻¹]	Daily waste generation per capita	1.00
p [−]	Number of citizens in the area under study	8345
W_comm [kg]	Daily waste generation in commercial activities	3655
W_tot [kg]	Total amount of waste generated per week	84,000
W_mix [kg]	Total amount of waste collected as mixed per week	84,000
N_CPA [−]	Number of CPs	88
a_wb [−]	Average number of waste bin per CP	2
ρ_mix[kg·m^–3]	Overall density of mixed waste	450
Waste collection parameters
v_c [m³]	Vehicle capacity	16
U_c [−]	Vehicle utilisation factor	0.9
F_c [−]	Frequency of waste collection per week	7
N_CP [−]	Number of CPs that a vehicle can service before it is filled to capacity	47
T_I [min]	Length of time taken for Stage I	10
D_I [km]	Distance from garage to service area	5
v_I [km h⁻¹]	Average speed of Stage I	30
D_II [km]	Average distance between two CP	0.2
T_{II d} [min]	Average time taken for driving between two CPs	2
v_II [km h⁻¹]	Average speed of Stage II	10
T_{II l} [min]	Loading time at a CP	2
D_III [km]	Distance from last CP to disposal facility	4
v_III [km h⁻¹]	Average speed of Stage III	20
T_III [min]	Length of time taken for Stage III	12
D_IV [km]	Distance from disposal facility to garage	3.6
v_IV [km h⁻¹]	Average speed of Stage IV	30
T_IV [min]	Length of time taken for Stage IV	7
T_C [min]	Length of time taken for one collection trip	190
N_WH [h]	Number of work hours per shift	8
T_L [min]	Lunch period	30
T_B [min]	Break period	15
N_CT [−]	Number of collection trips that a vehicle can make during one work shift	2.25
N_CTA [−]	Number of collection vehicle trips needed for service area	1.94
N_CTA [−]	Number of collection vehicle trips needed for service area	Round: 2
N_WD [−]	Number of workdays per week	7
N_V [−]	Number of vehicles needed to service all CPs	1
Costs of waste collection
C_v [€]	Unit price of vehicle	100,000
e [−]	Administrative rate	0.12
L [year]	Economic life of vehicle	10
i [−]	Yearly discount rate	0.085
CRF	Capital recovery factor	0.1524
C_cv [€ vechicle⁻¹year⁻¹]	Capital cost of vehicle amortised over its economic life	17,070
C_wb [€]	Unit price of waste bin	300
L_wb [year]	Economic life of waste bin	10
C_vwb [€ bin⁻¹year⁻¹]	Capital cost of waste bin amortised over its economic life	51.21
CRF_wh [−]	Capital recovery factor	0.1524
a [−]	Fringe benefit rates	0.24
b_w [−]	Backup rate for collection workers	0.05
S_w [€h⁻¹]	Salary rate per hour per worker	2
N_w [−]	Number of workers per vehicle	2
S_d [€h⁻¹]	Driver salary rate per hour	2.5
W_h [−]	Working hours per day	8
C [€ vechicle⁻¹year⁻¹]	Annual vehicle operation and maintenance costs	15,000
d [€ worker⁻¹year⁻¹]	Other expenses	500
C_ov [€ vechicle⁻¹year⁻¹]	Operating cost of collection vehicle	53,075
b_v [−]	Backup rate for collection vehicle	0.1
C_vy [€ year⁻¹]	Annual cost of collection per vehicle	71,852
C_ta [€ year⁻¹]	Total cost of waste collection per year	71,852
C_cp [€ year⁻¹]	Cost of waste collection per CP per year	816.5
C_t [€ t⁻¹]	Cost of waste collection per tonne	16.4

CP: collection point.

Values for parameters that are the results of calculations are shown in italic.

The first part, named ‘waste parameters’, refers to the waste quantity that should be collected. The value that ought to be entered into this part is number of citizens that live in area under study and daily waste generation per capita. Since the waste from the commercial sector is also taken into account, the amount of this type of waste is also calculated in the separate module of the Excel worksheets and it is not shown because it would expand the Table 1 too much and make it unsuitable to present. The module calculates the amount of waste according to equation (3), so all input data for calculation must be given.

Another added module to this table consistes of waste composition as well as density of each waste fraction. These are also input data for further calculation. Based on the waste composition and the degree of compaction for each component, overall density of mixed waste $(ρ_{m i x})$ is calculated. The data about morphological composition of solid waste, as well as waste quantity per capita, were taken from previous research conducted in the city of Kragujevac (Vujic et al., 2010). The average morphological composition of waste for the area under study is shown in Table 2.

Table 2.

Composition of waste in the study area (as a percentage of total category weight).

Waste category
Garden waste	6.23
Other biodegradable waste	27.09
Paper	12.78
Glass	5.92
Cardboard	15.15
Waxed cardboard	1.45
Aluminium-coated cardboard	1.76
Metal packaging and other	0.80
Metal – aluminium cans	1.41
Plastic packaging waste	5.38
Plastic bags	9.75
Hard plastics	3.22
Textiles	1.56
Leather	0
Nappies	2.47
Fine waste particles	5.03

The second part of Table 1 refers to waste collection parameters. The input data for this part of the table, such as vehicle characteristics, number of crew members, number of work hours per shift, break time periods and collection frequency, were obtained from the utility company. Data regarding distance and time for waste collection during different waste collection and transport stages were collected during field work and by using GPS equipment. As shown in Table 1, the results of calculations indicate that a vehicle can service 47 CPs before it is filled up to capacity, while total time for one collection trip is 190 minutes. Hereafter it can be seen that a vehicle can make 2.25 collection trips per work shift and a number of collection trips to service the whole area under study is 1.94 (rounded to 2). For the given frequency of 7 days per week, according to calculation, one vehicle can service all CPs.

The third part of Table 1 refers to costs of waste collection. Input data for this part, such as salaries, annual vehicle operation and maintenance costs and working hours per day, is provided by the utility company. Economic life of vehicles and waste bins are adopted to be 10 years. Other values, such as vehicle and bin price, are taken as an average market price.

The results of the calculated cost indicate that the annual cost per collection vehicle per year is € 71,852, per CP € 816.5, while the cost per tonne of collected waste is € 16.4. It should be emphasised that this is the cost only of waste collection and transport to the landfill. These costs do not include any other costs such as landfilling costs or cost of administration staff.

Another scenario analysed within this article includes collection of dry recyclables separately from residues. For that purpose, a waste bin for recyclables (PET, cans, glass) should be placed at each CP in the area under study. In our calculation of costs for this scenario, we assume that 90% of total waste producers participate in waste separation and the average separation rate per recyclable is adopted to be 0.8. In accordance with waste composition, the amount of recyclables that should be collected is 13,440 kg each week or 699 t per year. The frequency of waste collection is once per week and a collection vehicle needs three trips to service the whole area. The calculation of waste collection costs was made using the same procedure as for the mixed waste collection, but because of length limitations of this article, only the final costs will be shown (Table 3).

Table 3.

Costs of collection of recyclable waste.

C_vy [€ year⁻¹]	Annual cost of collection per vehicle	13,770
C_ta [€ year⁻¹]	Total cost of waste collection per year	13,770
C_cp [€ year⁻¹]	Cost of waste collection per collection point per year	156.5
C_t [€ t⁻¹]	Cost of waste collection per tonne	19.7

Only one vehicle volume was purchased, therefore, all waste types are collected by the same type of vehicle. The price of € 19.7 per tonne of collected recyclables refers only to its collection. In order to determine economic viability of separate collection, other costs, such as cost separation of each fraction, preparation for market and administrative staff costs, should be included.

Scenario 2 includes collection of residues waste too. Since, the residues waste also should be collected each day, from the same CPs and from the same waste bins as in Scenario 1, the costs of waste collection per year and per CP would be the same as shown in Table 1, because the parameters of waste collection remain same. The costs per tonne of collected waste will be € 19.6, which is a higher price because there would be less waste collected compared with Scenario 1. By introducing the separate collection of recyclable waste in Scenario 2, the total price of waste collection per CP increases from € 816.5 to € 973, which presents a higher price by 19.2% compared with Scenario 1.

By using the developed mathematical model and management tool in Excel worksheets, the costs could be calculated for any scenario, for example for separate collection of organic waste. In that case the number of necessary waste bins as well as collection frequency should be determined, and after that the complete procedure shown above for calculating costs could be applied.

Conclusions

Waste collection and transport can generate up to 70% of the MSW system costs. Separate collection of recyclables or organic waste raises the costs of waste collection. The proper estimation and monitoring of waste collection costs are essential to define the most cost-effective waste collection system. This has become a critical issue for every MSW company aimed at providing their services in the best possible way.

This article presents a methodological proposal for calculating the costs of waste collection for different waste collection schemes. After developing the mathematical model, a management tool was implemented. Based on input data, such as waste quantity and composition, the number of waste bins, the location of CPs, the type of collection vehicle, crew, collection route, etc., the developed tool can calculate the time and costs of waste collection (per vehicle, CP or tonne of collected waste). This tool allows benchmarking and variance analysis, and it can be used as a fast method to determine time and costs of waste collection and it can be very useful for MSW management companies.

The tool was tested on a district in the city of Kragujevac for two different scenarios. The findings of the analysis carried out show that collection costs rise by 19.2% when introducing the separate collection of recyclable waste in Scenario 2, compared with Scenario 1.

The proposed methodology has a great level of flexibility and robustness and so may be equally implemented in any waste management service and in any company without sophisticated management systems. It may also be applied to the collection of any waste fraction in the case of separate collection and equally to any kind of waste collection, such as door-to-door or curbside collection.

The implementation of the method proposed in this study might offer benefits for local authorities to guide the setting of a solid waste tariff, and to increase the company’s productivity and cover the costs by finding the optimal solutions for waste collection for each part of the city. This could also decrease the waste fees paid by the citizens.

Footnotes

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by the Ministry of Education, Science and Technological Development of the Republic of Serbia [III42013].

References

Barlaz

Brill

Kaneko

. (1995) Integrated solid waste management: 1 Mathematical modelling. In: Proceedings of the 2nd congress on computing in civil engineering, New York, USA, 5–8 June 1995, pp.1150–1157. Atlanta, USA: ASCE.

Boskovic Jovicic (2015) Fast methodology to design the optimal collection point locations and number of waste bins: A case study. Waste Management & Research 33: 1094–1102.

Boskovic

Jovicic

Milasinovic

. (2013) Methodology for reduction of GHG emission from municipal waste collection and transport. International Journal for Quality Research 7: 641–652

Council Directive 1999/31/EC of 26 April 1999 on the landfill waste. Official Journal of the European Communities, L 182/1–19, 16.07.1999.

Curtis

Dumas

(2000) A spreadsheet process model for analysis of costs and life-cycle inventory parameters associated with collection of municipal solid waste. North Carolina State University, USA.

D’Onza

Greco

Allegrini

(2016) Full cost accounting in the analysis of separated waste collection efficiency: A methodological proposal. Journal of Environmental Management 167: 59–65.

Greco

Allegrini

Del Lungo

. (2015) Drivers of solid waste collection. Empirical evidence from Italy. Journal of Cleaner Production 106: 364–371.

Huang

Pan

Kao

(2011) Performance assessment for municipal solid waste collection in Taiwan. Journal of Environmental Management 92: 1277–1283.

Jacobsen

Buysse

Gellynck

(2012) Cost comparation between private and public collection of residual household waste: Multiple case studies in the Flemish region of Belgium. Waste Management 33: 3–11.

10.

Jovicic

Makojevic

Boskovic

. (2015) Performance analysis of solid waste management companies in the Republic of Serbia. In: Proceedings of 9th international quality conference, Kragujevac, Serbia, 4–5 June 2015, pp.119–126. Kragujevac, Serbia: Center for Quality, Faculty of Engineering, University of Kragujevac.

11.

Karagiannidis

Xirogiannopoulou

Tchobanoglous

(2008) Full cost accounting as a tool for the financial assessment of Pay-As-You-Throw schemes: A case study for the Panorama municipality, Greece. Waste Management 28: 2801–2808.

12.

Larsen

Merrild

Møller

. (2010) Waste collection systems for recyclables: An environmental and economic assessment for the municipality of Aarhus (Denmark). Waste Management 30: 744–754.

13.

Official Gazette of RS 92/10, Regulation on waste disposal on landfills, 2010.

14.

Pires

Martinho

Chang

(2011) Solid waste management in European countries: A review of systems analysis techniques. Journal of Environmental Management 92: 1033–1050.

15.

Serbian Environmental Protection Agency (SEPA) (2014) State of the Environment in the Republic of Serbia 2014 Report [in Serbian]. Available at: http://www.sepa.gov.rs/download/Izvestaj2014.pdf (accessed 20 November 2015).

16.

Stanisavljevic

Ubavin

Batinic

. (2012) Methane emissions from landfills in Serbia and potential mitigation strategies: A case study. Waste Management & Research 30: 1095–1103.

17.

Tavares

Zsigraiova

Semiao

. (2009) Optimisation of MSW collection routes for minimum fuel consumption using 3D GIS modelling. Waste Management 29: 1176–1185.

18.

United States Environmental Protection Agency (1997) Full Cost Accounting for Municipal Solid Waste Management: A Handbook. Available at: http://www.epa.gov (accessed 25 December 2015).

19.

Vujic

Jovicic

Redzic

. (2010) A fast method for the analysis of municipal solid waste in developing countries – case study of Serbia. Environmental Engineering and Management Journal 9: 1021–1029.