Evaluation of operational,economic,and environmental performance of mixed and selective collection of municipal solid waste: Porto case study

Abstract

This article describes an accurate methodology for an operational, economic, and environmental assessment of municipal solid waste collection. The proposed methodological tool uses key performance indicators to evaluate independent operational and economic efficiency and performance of municipal solid waste collection practices. These key performance indicators are then used in life cycle inventories and life cycle impact assessment. Finally, the life cycle assessment environmental profiles provide the environmental assessment.

We also report a successful application of this tool through a case study in the Portuguese city of Porto. Preliminary results demonstrate the applicability of the methodological tool to real cases. Some of the findings focus a significant difference between average mixed and selective collection effective distance (2.14 km t^-1; 16.12 km t^-1), fuel consumption (3.96 L t^-1; 15.37 L t^-1), crew productivity (0.98 t h^-1 worker^-1; 0.23 t h^-1 worker^-1), cost (45.90 € t^-1; 241.20 € t^-1), and global warming impact (19.95 kg CO_2eq t^-1; 57.47 kg CO_2eq t^-1).

Preliminary results consistently indicate: (a) higher global performance of mixed collection as compared with selective collection; (b) dependency of collection performance, even in urban areas, on the waste generation rate and density; (c) the decline of selective collection performances with decreasing source-separated material density and recycling collection rate; and (d) that the main threats to collection route efficiency are the extensive collection distances, high fuel consumption vehicles, and reduced crew productivity.

Keywords

Municipal solid waste waste collection performance indicators life cycle assessment performance assessment

Introduction

Municipal solid waste (MSW) management includes different kinds of issues (environmental, technological, economic, legislative, and social, among others) and problems (waste generation, collection, transport, treatment, and disposal) (Anghinolfi et al., 2013). MSW organisations are open systems where energy, resources, and financial resources enter, and emissions and raw materials exit through the system boundaries.

Collection and transport are the contact points between waste generators and waste management systems (Kanchanabhan et al., 2011). Both of these factors vary significantly over time and involve important operational problems and activities (Faccio et al., 2011). Collection represents 50% to 70% of the waste management cost, and fuel is often the greatest expense (Nguyen and Wilson, 2010). Collection is the stage with the most significant impact on the public since inefficient waste collection and/or insufficient numbers of containers or collection points results in serious inconveniences. Other issues range from visual impacts, odours, and potential public health problems. Even regular collection can be a source of trouble from a citizen’s point of view in the forms of night-time noise or daytime traffic congestion.

MSW selective collection (paper/cardboard, glass, and lightweight packaging) is accepted as a way to preserve natural resources and the environment. Indeed, European legislation (Directive 2008/98/EC) imposes continually higher recycling goals and recommends a growing number of source-separated materials. However, selective collection requires increases in the number of collection routes and considerable logistical support, such as vehicles, containers, human resources, and fuel. Selective collection is particularly critical in areas of low population densities and high dispersion. Indeed, Larsen et al. (2009) and Eisted et al. (2009) detected a proportional correlation between diesel consumption and waste or population density. They highlighted highest diesel consumption per metric tonne of waste collected, in areas with long distances and small amounts of waste. Eisted et al. (2009) also noted higher fuel consumption for the collection of materials with low density. Therefore, MSW collection operators and decision-makers need effective methodologies and tools to support management options under uncertain and complex operational issues, such as population, area, costs, equipment, and human resources. Furthermore, the forthcoming MSW collection management tools should simultaneously consider the operational, economic, and environmental performance efficiency.

Operational and economic performance can be supported by key performance indicators (IDs), as is a regular practice. Often the entities have to report IDs to a regulatory authority. In this respect, Del Borghi et al. (2009) claims that collection efficiency can be accessed through some IDs, such as collection frequency, distance to disposal site, and total cost.

MSW environmental assessment could be implemented through the life cycle assessment (LCA) (Anghinolfi et al., 2013; Del Borghi et al., 2009; Ekvall et al., 2007; Eriksson et al., 2005; Kirkeby et al., 2006; Obersteiner et al., 2007; Schmidt and Pahl-Wostl, 2007; Solano et al., 2002a, 2002b; Winkler and Bilitewski, 2007) and its inventory data are usually collected specifically for this purpose. However, Lo et al. (2005) raised some concerns because the MSW environmental impacts using LCA are mostly based on proposed scenarios, and their final results are mainly grounded on uncertain fates.

Furthermore, Winkler and Bilitewski (2007) state that efforts must be made to increase the transparency of data and to reduce the degree of uncertainty in these models in order to obtain more robust results. Traditional LCA models applied to MSW analysis have great detail in waste treatment, but do not handle the collection stage with the same level of detail, most likely because this phase requires a lot of information.

In this context, the MSW collection and planning requires robust operational, economic, and environmental solutions. Given the above, the objective of this article is to analyse, in detail, the MSW collection stage, to contribute to the existing models, and to make a comparison between selective and mixed collection methods. This work also shows that it is possible to use IDs in the LCA inventory by utilising the information made available by the numerous entities that regularly calculate IDs. On the other end, IDs are normalised, reduced to the functional unit (e.g. by amount collected). Note also that the ID can be calculated by route collection, type of container, and type of waste, among others. So, in this way, it is possible to adjust the level of detail included in the environmental analysis. The tool was tested through a case study in the Portuguese city of Porto. For this purpose, we used data from mixed and selective collection obtained during 12 weeks of monitoring services. The main disadvantage of this methodology is the incomplete analysis of the MSW management system, namely the study of the collection stage.

Methodology

The proposed MSW collection performance evaluation methodology ensures operational, economic, and environmental assessment. It is based on a reduced number of IDs to achieve independent operational and economic efficiency, and uses the LCA for environmental assessment (Figure 1). The life cycle inventories and impact assessment methods provide environmental assessments through impact categories.

This performance assessment approach is normalised by the weight of collected material and the IDs can be applied to each collection route allowing their individual performance evaluation. Thereafter, performance results can be grouped by type of container, vehicle, and waste (e.g. mixed, paper, glass, and light packaging).

Successful performance evaluation demands logistic, operational, economic, and environmental information. Logistic and operational information requires the acquisition of individual collection route data. The vehicle crew ends an itinerary after all the assigned containers have been loaded or when the vehicle reaches its capacity. Vehicle drivers are responsible for assembling operational data by filling service forms with information regarding route identification, crew size, waste type (mixed, paper/cardboard, glass, and lightweight packaging), weight of each collection route load, fuel consumption, number of loaded containers, container type, and waiting time at the disposal site. Additionally, they must note the distances and times travelled between the itinerary start to the first collection point, the first container to the last container, the last container to the disposal site, and the disposal site back to the parking area. The fuel consumption measure requires that each vehicle driver fill the fuel tank at the beginning and end of service.

Economic information used to evaluate cost assignment includes investment and capital data in all stages of the waste collection process. This evaluation requires data collection of expenditures related to processing technology, building and equipment, capital requirements, commitment interest, depreciation and amortisation, insurance, and taxes. Also, ordinary operational expenses are required: salaries, vehicle repair and maintenance, electricity, office and additional costs, fuel costs, accessory and operating material costs (e.g. containers and bags), among others.

Operational and economic information is then processed and assigned to each collection route. This collection database provides the IDs support variables (Table 1).

Table 1.

Performance indicators support variables.

Variable	Designation	Unit	Definition
Dt	Total collection distance	km	Travel distance from parking to first collection point, distance from the first to last collection point, travel from the last collection point to disposal site (landfill, recovery facility, transfer station) where the collection vehicle is unloaded (including waiting time), and finally, the travel distance from disposal site to parking or restart of new collection.
De	Effective collection distance	km	Distance from the first to last collection point.
Tt	Total collection time	h	Time spent on travel from parking to first collection point, time from the first to last collection point, time from last collection point to disposal site (landfill, recovery facility, transfer station) where the collection vehicle is unloaded (including waiting time), and time from disposal site to parking or the start of a new collection.
Te	Effective collection time	h	Time from the first to last collection point.
Fc	Fuel consumption	L	Fuel consumption at each collection route.
Cn	Crew members	worker	Number of crew members.
MSWc	Municipal solid waste	t	Amount of MSW collected at each collection route.
Cc	Collection cost	€	Annual investment, capital needs (yearly basis) and operational collection costs (container placing and maintenance, MSW collection, and transport to the disposal site).
Inh	Municipality’s inhabitants	inhab.	Municipality’s inhabitants.
Hc	Household collection contracts	household	Municipality’s household collection contracts.

MSW: municipal solid waste.

Then, support variables are normalised by weight collected, setting the nine proposed IDs (Table 2). They are grouped into two groups: operational indicators (total collection distance, effective collection distance, total collection time, effective collection time, fuel consumption, and crew productivity) and economic indicators (collection cost, inhabitant cost, and household cost).

Table 2.

Performance indicators description.

	Designation	Unit	Definition
Operational indicators
$IDt = \frac{Dt}{MSWc}$	Total collection distance	Km t^-1	Distance from departure to parking entrance at each collection route.
$IDe = \frac{De}{MSWc}$	Effective collection distance	km t^-1	Distance to load the collection vehicle on each collection route. This indicator measure begins with the loading of the first collection point and ends when the last container is loaded.
$ITe = \frac{Te}{MSWc}$	Total collection time	h t^-1	Time spent from departure to parking entrance at each collection route.
$ITe = \frac{Te}{MSWc}$	Effective collection time	h t^-1	Time to load the collection vehicle on each collection route. This indicator measure begins with the stop to load at the first collection point and ends when the last container is loaded.
$IFc = \frac{Fc}{MSWc}$	Fuel consumption	L t^-1	Amount of fuel used per metric tonne of MSW collected on each collection route.
$IProd = \frac{MSWc}{(Te \times Cn)}$	Crew productivity	t h^-1 worker^-1	Crew member’s effort to collect the total amount of MSW on each collection route.
Economic indicators
$ICc = \frac{Cc}{MSWc}$	Collection cost	€ t^-1	Economic effort to collect municipality’s MSW on each collection route.
$IInhc = \frac{Cc}{Inh}$	Inhabitant cost	€ inhab.^-1	Economic effort to collect municipality’s per capita solid waste.
$IHc = \frac{Cc}{Hc}$	Household cost	€ household^-1	Economic effort to collect each household collection contract.

MSW: municipal solid waste.

The MSW collection environmental assessment is carried out through the LCA methodology. The functional unit is the collection of one metric tonne of MSW generated. The reference time unit is the year, and the municipality area is coincident with the geographical system boundaries. Since this study focused on the MSW collection stage, the LCA system boundaries included storage of waste in containers (pre-collection stage), collection and transport to end-life disposal (landfill or incineration), and transport to a material recovery facility (MRF). The study excluded subsequent processes, such as landfilling (mixed collection), MRF processing, transportation to the recycling facilities, and the obtained recyclable fractions (selective collection). Additionally, the citizens’ transportation of waste to containers was outside the scope of this study.

An exhaustive data inventory (i.e. on energy and resource consumption) was conducted for each collection activity and supporting facilities. The ecoinvent database (Frischknecht and Rebitzer, 2005; Swiss Centre for Life Cycle Inventories, 2011) was used to obtain the inventory data for the materials involved in this study; namely diesel fuel, electricity, and water. In order to adapt an electricity consumption inventory to Portuguese reality, the ecoinvent dataset #631 was used. This database is related to the Portuguese average electricity mix (57% oil, 15% renewables, 13% solid fuels, 13% gas, and 2% other).

Diesel fuel data was obtained directly from the IFc performance indicator, fuel consumption (L t^-1) and the corresponding transport from the performance indicator IDt, total collection distance (km t^-1), calculated for every collection route. However, regarding electricity and water consumed in the collection phase, the corresponding IDs were not calculated because they were not considered relevant to operational assessment. Thus, in this particular case, we considered the total annual consumption of electricity and water and divided it by the total amount of waste produced. Using this methodology, electricity and water were also reduced to functional units, kWh t^-1 and m³ t^-1, respectively.

The main materials and capacities of the most common types of containers are shown in Table 3. The LCA inventory data regarding containers was compiled from Bovea et al. (2010) and Rives et al. (2010). The volume and number of street-side and drop-off containers were registered and used to calculate the number of containers needed to satisfy the functional unit.

Table 3.

Main components of containers in MSW collection systems.

Container model	Capacity (m³ cont^-1)	Component	Material	Quantity of material (kg cont^-1)
Street-side	1.1	Body + inlet door + four wheels	HDPE	59.0
		Body + inlet door + four wheels	Steel	4.0
			Rubber	1.5
			Aluminium	0.5
	0.8	Body + inlet door + four wheels	HDPE	42.0
		Body + inlet door + four wheels	Steel	4.0
			Rubber	1.5
			Aluminium	0.5
Drop-off	5.0	Body	HDPE	140.0
		Cover	HDPE	18.5
		Auxiliary parts	HDPE	8.0
		Lifting bag	PP	20.0
		Disposable bag	HDPE	1.9
	2.5	Body + lifting system	HDPE	155.0
			Steel	4.5

HDPE: High Density Polyethylene; PP: Polypropylene.

The proposed environmental assessment methodology only includes the LCA’s classification and characterisation stages. The adopted impact assessment method is the CML 2 baseline 2000 method (Guinée et al., 2001). The impact categories considered in the study are global warming (GWP100), ozone layer depletion (ODP), acidification (AP), and human toxicity (HTP). Specific software based on Microsoft Excel^© spreadsheets were developed to speed up overall assessment.

Preliminary results and discussion

This methodological tool was further developed for mixed and source-separated (paper/cardboard, glass, and lightweight packaging) collection routes from the Portuguese municipality of Porto. Municipality background information reports 263,131 inhabitants (5787 inhab. km^-2) and 152,000 household collection contracts. MSW generation was 124,968 metric tonnes: 117,815 metric tonnes from mixed collection and the remaining 7153 metric tonnes from selective collection as follows: 40.26% glass, 47.06% paper/cardboard, and 12.68% light packaging (High Density Polyethylene (HDPE), Low Density Polyethylene (LDPE), Polyethylene Terephthalate (PET), liquid packaging board, and ferrous and non-ferrous metals). Table 4 shows the composition (% by weight) and densities within containers (kg m^-3) of selective fractions and mixed waste obtained from the Porto municipality.

Table 4.

Porto: average waste composition and density in container.

Fraction		Composition (% weight)	Density in container (kg m^-3)
Fraction		Composition (% weight)	Min.	Max.	Average
Light packaging	Plastic	14.8	45	79	67
	Metal	1.5	128	194
Paper/cardboard		18.5	57	96	86
Glass		4.6	169	213	194
Mixed waste		60.6	85	276	135

The mixed collection system carried out through 55 routes with different collection frequencies: 30 daily routes and 25 weekly routes. Most daily routes (23) use street-side containers (0.8 or 1.1 m³) and the remaining seven routes use drop-off containers (5 m³). All weekly routes use street-side containers. The street-side container collection routes apply rear loading compacting vehicles (15–20 m³). A crane vehicle is used to empty the drop-off containers. The selective collection system was carried out through 10 routes for each recyclable material (paper/cardboard, glass, and lightweight packaging) with various collection frequencies (1–4 days per week). This system used drop-off containers (2.5 m³) collected through crane loading vehicles (20 m³). Both mixed and selective crews employ three workers from Monday to Saturday, eight hours each day. The disposal site is 5 km outside the municipal limits.

The methodological tool application to Porto’s collection system proves the method’s capacity to perform operational, economic, and environmental assessment, and subsequent comparison between mixed and selective collection. All indicators and environmental profiles reflect the average of 2312 observations from mixed collection routes and 720 from selective collection, with 240 observations for each recyclable material (paper/cardboard, glass, and lightweight packaging).

The preliminary results presented in Table 5 demonstrate the methodological tool applicability to real cases. Preliminary results show a consistent difference between mixed and selective collection. Generally, the results of mixed collection reflect a better performance as they have higher productivity with lower fuel consumption and smaller collection distances.

Table 5.

Operational assessment of MSW collection system.

Performance indicator	Designation	Unit	Mixed collection	Selective collection
Total collection distance	IDt	km t^-1	6.31	21.08
Effective collection distance	IDe	km t^-1	2.14	16.12
Total collection time	ITt	h t^-1	0.46	5.91
Effective collection time	ITe	h t^-1	0.34	4.89
Fuel consumption	IFc	L t^-1	3.96	15.37
Crew productivity	IProd	t h^-1 worker^-1	0.98	0.23

The ID values show a 34% effective collection distance rate and 70% effective collection time. The selective, effective, collection distance rate was 76% and the effective collection time was 83%. The results lead to the conclusion that, in the case of mixed collection, the distances travelled from the last container to the disposal site (landfill, recovery facility, and transfer station) were too far. Thus, it is recommended that the itinerary be amended. Further, data demonstrates that travel along selective collection routes is much steadier than on mixed collection routes.

The average diesel consumption from mixed collection (3.96 L t^–1) fits into the range proposed by Larsen et al. (2009) (1.4–3.6 L t^–1). Selective collection values (15.37 L t^–1) are 4.3 times greater than the maximum limit of the range. The key performance indicator results show a positive correlation between fuel consumption and effective time and distance travelled. This correlation is in agreement with the results reported by Larsen et al. (2009).

Operational results show an opposite correlation between distance or time travelled and crew productivity. The results also indicate increased fuel consumption from glass to packages, which is in agreement with Eisted et al. (2009), who stated that fuel consumption is higher for materials with low density.

Selective collection operational assessment (Table 6) indicates lower crew productivity than in mixed collection as a result of the reduced effective distance: 45% in paper (13.37 km t^-1), 50% for glass (8.13 km t^-1), and 42% in packaging (40.05 km t^-1). Among the selective collections packaging collection (46.32 L t^-1) was responsible for the highest fuel consumption. The average fuel consumption in the collection of glass and paper ranged from 8.39 to 14.22 L t^-1, which exceeds the range proposed by Larsen et al. (2009) (3.4–6.6 L t^-1). The selective collection productivity ranged from 0.07 t h^-1 worker^-1 for packaging to 0.37 t h^-1 worker^-1 for glass.

Table 6.

Operational assessment of the selective collection system.

Operational indicators	Designation	Units	Paper	Glass	Light packaging
Total collection distance	IDt	km t^-1	29.43	16.37	94.48
Effective collection distance	IDe	km t^-1	13.37	8.13	40.05
Total collection time	ITt	h t^-1	2.54	1.27	7.83
Effective collection time	ITe	h t^-1	1.84	0.99	5.71
Fuel consumption	IFc	L t^-1	14.22	8.39	46.32
Crew productivity	IProd	t h^-1 worker^-1	0.21	0.37	0.07

Appropriate comparative analysis between selective (sc) and mixed collection (mc) performance was carried out by the effectiveness rate (ER_Pi) (equation (1)), which is the quotient between the proposed key IDs ( Pi ) for both collection types (Figure 2):

{ER}_{p i} = \frac{P i_{s c}}{P i_{m c}}

(1)

Key performance indicators report better operational effectiveness rate (ER_pi) in mixed routes (Figure 2). It was noted that the effective time to collect one metric tonne of recycled waste was 14.4 times higher than in mixed collection (ER_pi = 14.4) and the effective length was 7.5 times higher (ER_IDe = 7.59). In the category of fuel consumption 3.9 times more diesel per metric tonne (ER_IFc = 3.9) was spent during selective collection (Figure 2(a)). The crew’s productivity range ER_IProd was 0.1 for packaging and 0.4 for glass. Finally, selective collection performance was constrained by the material’s average density. The overall performance of selective collection was found to be consistently lower than that of mixed collection. For example IDe of packages is 18.7 times higher than mixed collection (ER_IDe = 18.7).

The economic IDs (Table 7) show the unitary collection costs and the inhabitant and household contract unitary costs. The economic IDs take into account the expenses presented in Table 8, which highlight that Porto’s major expenses were labour costs and vehicle maintenance.

Table 7.

Economic assessment of the waste collection system.

Economic indicators	Units	Mixed collection	Selective collection
Collection cost	€ t^-1	45.90	241.20
Inhabitant cost	€ inhab.^-1 year^-1	20.52	6.50
Household cost	€ household^-1 year^-1	35.34	11.28

Table 8.

Waste collection expenses (%).

Expenses	Mixed collection	Selective collection
Overheads	3.19%	2.99%
Vehicles maintenance	21.17%	35.87%
Vehicles acquisition	10.84%	6.40%
Fuel	8.56%	5.17%
Salary costs	54.01%	47.60%
Facilities	2.23%	1.97%
Total (%)	100%	100%

The major cost differences between selective and mixed collection were: per capita effectiveness costs (ER_IInhac = 0.32) and household effectiveness costs (ER_IHc = 0.32). These differences result from the need to ensure the availability and accessibility of the service in terms of selective collection containers, workers, vehicles, and administration despite the reduced rate of separate collection (5.7%). However, despite the low source-separated collection rate, it is necessary to ensure an infrastructure that guarantees access to this service. This option leads to higher costs in selective collection. Finally, once again, the economic IDs show an increase in the unit costs of glass collecting in comparison with collection of packaging, which supports the idea that higher collection costs correspond to recycling lower density materials.

The MSW collection environmental assessments were carried out through LCA and normalised by weight collected. The collection supporting facility’s LCA inventory data were collected directly from the Porto municipality. Key performance indicators IFc and IDt support the environmental assessment from collection routes and waste type. As previously mentioned, the number and volume of street-side and drop-off containers were registered and used to calculate the number of containers needed to satisfy the functional unit (FU) (Table 9). Electricity and tap water consumption were collected and assigned to the FU (Table 10).

Table 9.

Containers capacity and number of containers by FU.

Fraction	Container capacity (m³ cont^-1)	Containers (nº t^-1)
Mixed waste	0.8	12.3
	1.1	9.0
	5.0	2.0
Paper/cardboard	2.5	6.2
Glass	2.5	2.7
Light packaging	2.5	8.0

Table 10.

Electricity and tap water consumed in collection supporting facilities.

Collection	Electricity (kWh t^-1)	Water (m³ t^-1)
Mixed	0.424	1.226
Paper/cardboard	0.010	0.030
Glass	0.012	0.035
Light packaging	0.003	0.009

The environmental assessment, reported by waste type, according to the system boundaries and assumptions, was consistent with the finding that mixed collection yields better operational performance (Table 11). Light packaging collection results show a poor environmental performance in line with the equally poor operating performance.

Table 11.

Environmental assessment of waste collection system.

Impact category	Designation	Unit	Mixed collection	Selective collection	Paper	Glass	Light packaging
Global warming	GWP100	kg CO_2eq t^-1	19.95	57.47	51.17	28.59	174.39
Ozone layer depletion	ODP	kg CFC-11_eq t^-1	1.02 × 10^-5	3.49 × 10^-5	3.13 × 10^-5	1.69 × 10^-5	0.10 × 10^-5
Human toxicity	HTP	kg 1,4-DCB_eq t^-1	1.48	4.64	4.14	2.25	14.18
Acidification	AP	kg SO_2eq t^-1	0.13	0.42	0.38	0.21	1.28

Using Eisted et al. (2009) GWP100 values for mixed (5–5.5 kg CO_2eq t^-1) and selective paper collection (6–11 kg CO_2eq t^-1) as reference factors, the results for the Porto case study are 3.6 times higher (19.95 kg CO_2eq t^-1) on mixed collection and 4.65 times on paper-separated collection (51.17 kg CO_2eq t^-1).

Suitable comparative analysis between selective (sc) and mixed collection (mc) environmental performance was carried out by the effectiveness rate (ER_ep) (equation (2)), which is calculated by the quotient between the adopted environmental impact categories (Ic) of both collection types (Figure 3).

E R_{e p} = \frac{I c_{s c}}{I c_{m c}}

(2)

Environmental effectiveness rate analysis (Figure 3) validates the better performance of mixed collection when contrasted with selective collection, for all the environmental categories. The environmental results display lower performance for selective collection, ranging from 2.9 to 3.4 for ER_GWP and ER_ODP, respectively (Figure 3(a)). This may be owing to a short recycling collection rate.

The highest ER_ep values were found in packaging collection and the lowest in glass collection. This shows that environmental impact is higher for packaging than for paper and glass, which is explained by the density of the recycled materials. It is noted that the environmental impact can be 10 times higher than mixed collection in the case of ODP ER_ODP = 10.4 (Figure 3(b)).

The city of Porto selective collection assessment revealed a reduced overall performance as a result of undeveloped operational, economic, and environmental efficiency when compared with the mixed collection method. Inefficient itineraries, low separated collection rate, uncontrolled labour and vehicle maintenance costs, and reduced environmental performance all contributed to the reduced overall performance of the selective collection method.

The methodological tool explored in this study reveals the need for mandatory guidelines to improve the municipality’s operational, economic, and environmental performance. Their major task is to develop the current collection system, which is based on strong mixed collection targets and reserves selective collection to a complementary role. Selective collection performance improvement is possible through three simultaneous approaches: recycling collection rate growth, suppression of inefficient collection routes, and the progressive reduction of crews, containers, and vehicles.

Conclusions

The proposed methodology for MSW collection performance evaluation is a useful tool for simultaneous operational, economic, and environmental assessment. The IDs can be used independently on operational and economic assessment, as well as basic support in life cycle inventories and life cycle impact assessments.

This methodology applies comparable procedures on database assembly and is supported by result verification and validation routines. The methodological tool supports the municipality’s sustainability efforts and provides internal and external benchmarking routines.

The tool was successfully implemented on a daily basis and Porto’s preliminary results indicate:

higher global performance on mixed collection than in selective collection;

dependence of collection performance, even in urban areas, on the waste generation rate and density;

decline of selective collection performance with decreasing material density and collection rate;

the main threats to collection route efficiency are the extensive collection distances, vehicles’ high fuel consumption, and reduced crew productivity.

This methodological tool is applicable to other systems in different regions and countries. Future developments may set a more restricted group of appropriate IDs and/or expand them to include waste treatment and disposal. In addition, further developments should focus on waste collection comparative assessment among high and low density areas.

Footnotes

Declaration of conflicting interests

The authors declare that there is no conflict of interest.

Funding

The authors would like to thank the municipality of Porto for all technical support and promptness while developing the current work.

References

Anghinolfi

Paolucci

Robba

Taramasso

(2013) A dynamic optimization model for solid waste recycling. Waste Management 33(2): 287–296.

Bovea

Ibáñez-Forés

Gallardo

Colomer-Mendoza

(2010) Environmental assessment of alternative municipal solid waste management strategies. A Spanish case study. Journal of Waste Management 30(11): 2383–2395.

Del Borghi

Gallo

Del Borghi

(2009) A survey of life cycle approaches in waste management. International Journal of Life Cycle Assessment 14(7): 597–610.

Eisted

Larsen

Christensen

(2009) Collection, transfer and transport of waste: accounting of greenhouse gases and global warming contribution. Waste Management & Research 27: 738–745.

Ekvall

Assefa

Björklund

. (2007) What life-cycle assessment does and does not do in assessments of waste management. Waste Management 27(8): 989–996.

Eriksson

Carlsson Reich

Frostell

. (2005) Municipal solid waste management from a systems perspective. Journal of Cleaner Production 13: 241–252.

European Parliament and Council of the European Union (2008) Council Directive 2008/98/EC on waste. Official Journal of the European Communities, Brussels.

Faccio

Persona

Zanin

(2011) Waste collection multi objective model with real time traceability data. Waste Management 31(12): 2391–2405.

Frischknecht

Rebitzer

(2005) The ecoinvent database system: A comprehensive web-based LCA database. Journal of Cleaner Production 13(13–14): 1337–1343.

10.

Guinée

Gorree

Heijungs

. (2001) Life Cycle Assessment; Operational Guide to the ISO Standards; Parts 1 and 2. Ministry of Housing, Spatial Planning and Environment (VROM) and Centre of Environmental Science (CML), Den Haag and Leiden, Netherland.

11.

Kanchanabhan

Mohaideen

Srinivasan

Kalyana Sundaram

(2011) Optimum municipal solid waste collection using geographical information system (GIS) and vehicle tracking for Pallavapuram municipality. Waste Management & Research 29(3): 323–339.

12.

Kirkeby

Birgisdottir

Hansen

. (2006) Environmental assessment of solid waste systems and technologies: EASEWASTE. Waste Management & Research 24: 3–15.

13.

Larsen

Vrgoc

Christensen

Lieberknecht

(2009) Diesel consumption in waste collection and transport and its environmental significance. Waste Management & Research 27: 652.

14.

S-L

(2005) Quantifying and reducing uncertainty in life cycle assessment using the Bayesian Monte Carlo method. Science of the Total Environment 340: 23–33.

15.

Nguyen

Wilson

(2010) Fuel consumption estimation for kerbside municipal solid waste (MSW) collection activities. Waste Management & Research 28(4): 289–297.

16.

Obersteiner

Binner

Mostbauer

Salhofer

(2007) Landfill modelling in LCA – a contribution based on empirical data. Waste Management 27(8): S58–S74.

17.

Rives

Rieradevall

Gabarrell

(2010) LCA comparison of container systems in municipal solid waste management. Waste Management 30(6): 949–957.

18.

Schmidt

Pahl-Wostl

(2007) Modelling biowaste flows for life-cycle assessment – calculation of the potential and collected weight of kitchen and garden waste. Journal of Industrial Ecology 11(1): 181–199.

19.

Solano

Ranjithan

Barlaz

Brill

(2002a) Life-cycle-based solid waste management. I: Model development. Journal of Environmental Engineering 128(10): 981–992.

20.

Solano

Dumas

Harrison

. (2002b) Life-cycle-based solid waste management. II: Illustrative applications. Journal of Environmental Engineering 128: 993–1005.

21.

Swiss Centre for Life Cycle Inventories (2011) Ecoinvent, V2.2, c/o Empa/ Technology & Society Lab (TSL), Lerchenfeldstrasse 5, 9014 St-Gallen, Switzerland.

22.

Winkler

Bilitewski

(2007) Comparative evaluation of life cycle assessment models for solid waste management. Waste Management 27(8): 1021–1031.