Resources, Conservation and Recycling 30 (2000) 111–133 www.elsevier.com/locate/resconrec

Strategic planning of municipal solid waste management Juha-Heikki Tanskanen * Finnish En6ironment Institute, PB 140, FIN-00251 Helsinki, Finland Received 12 July 1999; accepted 1 February 2000

Abstract A computer model was developed and applied for studying integrated municipal solid waste management (MSWM) in the Helsinki Metropolitan Area. The model is based on a method developed for analysing on-site collection systems of waste materials separated at the source for recovery. The aim of the Helsinki study was to find and analyse separation strategies fulfilling the recovery rate targets adopted for municipal solid waste in Finland, i.e. 50wt.% by the end of 2000 and 70wt.% by 2005. In the present situation (i.e. in 1995), the total recovery rate of 27wt.% was achieved in the region. The strategies studied were first based on source separation only, resulting in a highest recovery rate of 66wt.%. At the same time, the costs of MSWM increased by 41% compared to the year 1995. Next, a recovery rate of 74wt.% was attained by combining source separation with central sorting of mixed waste. As a result, the costs of MSWM increased by 30% compared to the present situation. In both of these strategies, the emissions caused by MSWM were generally reduced. The model developed proved to be a suitable tool for strategic planning of MSWM. Firstly, the analysis of collection systems helped to identify potential separation strategies and to calculate the amounts of materials collected for recovery. Secondly, modelling of MSWM systems made it possible to determine the effects of separation strategies on costs and emissions caused by the whole MSWM. The method and model developed can be also applied in other regions, municipalities and districts. © 2000 Elsevier Science B.V. All rights reserved. Keywords: Waste management; Municipal solid waste; Recovery rate; Separation; Materials; Costs; Emissions; Models

* Tel.: + 358-0-40300421; fax: + 358-0-40300491. E-mail address: [email protected] (J.-H. Tanskanen). 0921-3449/00/$ - see front matter © 2000 Elsevier Science B.V. All rights reserved. PII: S0921-3449(00)00056-2

112

J.-H. Tanskanen / Resources, Conser6ation and Recycling 30 (2000) 111–133

1. Introduction Integrated municipal solid waste management (MSWM) can be defined as the selection and application of suitable techniques, technologies and management programs to achieve waste management objectives and goals [1]. Computer models can be used as tools in the planning of integrated MSWM systems. During the past three decades, models have been developed in accordance with waste management objectives, especially waste minimization and emissions control. The reviews compiled by Gottinger [2] and MacDonald [3] show that early MSWM models developed during the 1960s and 1970s focused on studying individual functional elements, i.e. determining collection routes or facility locations, capacities or expansion patterns. In the 1980s, the focus was extended to cover MSWM on the system level, resulting in extended system boundaries. These models were mainly aimed at minimizing the costs of mixed waste management [2,4], and recycling was included in some of them more or less comprehensively [5,6]. In the 1990s, recycling has been extensively included in most models used for strategic planning of MSWM. Reduced system costs are the most common objective [7 – 14], but some models study MSWM from the point of view of the size and characteristics of waste streams [15,16] or their emissions [17]. In several strategic planning models, both costs and emissions of MSWM have been included in the study [18 – 22]. In some models, the whole life cycle of products has been studied instead of only the waste management system when searching for environmentally optimal waste management strategies [23,24]. Despite the development of strategic planning models, the descriptions of source separation strategies of recyclables are usually insufficient to enable calculation of the amounts of materials separately collected. The amount of a material separately collected in an area depends on two factors: (1) the coverage of a collection system applied and (2) the separation activity of waste producers, consisting of participation rate and separation efficiency. The coverage of a collection system is defined as the ratio of (a) the amount of a material produced in those properties where separate collection is available and (b) the amount of the material in question produced in all properties of the area. Participation rate is defined as the share of people providing sorted material to bins in those properties where this option is available. Separation efficiency is defined as the share of a material which is correctly separated by those participating in separation. In several strategic planning models, all of these factors have been ignored and the amounts of materials separated at the source are treated as input data [10,13,18–20]. In some models, the amounts of materials separated are calculated on the basis of participation rates and separation efficiencies [11,12]. However, the analysis of the coverages of collection systems has generally been excluded from strategic planning models. This paper presents the HMA (Helsinki Metropolitan Area) model developed for integrated analysis of recovery rates, costs and emissions of MSWM. The HMA model differs from most earlier models through a method developed to analyse the coverages of on-site collection systems of waste materials. Thus, the amounts of

J.-H. Tanskanen / Resources, Conser6ation and Recycling 30 (2000) 111–133

113

materials separately collected for recovery can be calculated on the basis of the characteristics of source separation strategies and area studied. Costs and emissions of MSWM are calculated on the basis of waste streams and corresponding unit costs and unit emissions. The method used in the HMA model can also be applied to study the effects of separation in other regions, municipalities and districts. The HMA model was applied in a case study performed in the Helsinki Metropolitan Area. The aims of the Helsinki study were: (1) to find separation strategies fulfilling the recovery rate targets adopted for municipal solid waste in Finland, i.e. 50wt.% by the end of 2000 and 70wt.% by 2005 [25] and (2) to determine the effects of these strategies on the costs and emissions of MSWM. Separation strategies were largely based on source separation according to the policy of the Helsinki Metropolitan Area Council. Recovery rate was determined as the share of waste which is separated and delivered to material or energy markets. Participation rates and separation efficiencies were expressed as separation activity, because of insufficient input data. The study covered all municipal solid waste from households and commercial premises. Wastes generated by e.g. construction and demolition activities as well as by waste water treatment plants were excluded from the study. The preliminary results of the case study have been presented by Tanskanen [26,27].

2. Materials and methods

2.1. The modelling concept The approach used in the HMA model can be divided into six stages (Fig. 1).

Fig. 1. Stages of the approach used in the HMA model.

114

J.-H. Tanskanen / Resources, Conser6ation and Recycling 30 (2000) 111–133

Firstly, potential separation strategies are formulated for recoverable waste materials on the basis of an analysis in which the coverages of different kinds of collection systems are determined. Waste producers are divided into groups, e.g. residential properties and commercial establishments, so that differences in the amount of materials produced can be taken into consideration when planning separation strategies. In addition to source separation, strategies may include central sorting of mixed waste. Secondly, the total recovery rate and the recovery rates of individual materials are calculated (Eqs. (1)–(7)). After the second stage, the separation strategies can be modified if the recovery level is too low.

 

R = %g Rg ×

sg 100

Rg =%i Rg,i ×

 

(1)

sg,i 100

(2) (3)

Rg,i =r1,g,i +r2,g,i r1,g,i =co,g,i ×

Po,g,i eo,g,i P e × + cd,g,i × d,g,i × d,g,i 100 100 100 100

100 ]co,g,i +cd,g,i r2,g,i =







(4)



n

(5)

e e ei P P × cx,g,i +cy,g,i × 1− o,g,i × o,g,i + cz,g,i × 1− d,g,i × d,g,i 100 100 100 100 100

(6) 100 ] cx,g,i +cy,g,i +cz,g,i

(7)

where cd,g,i is the coverage of drop-off centre collection of material i in waste producer group g (%); co,g,i, coverage of on-site collection of material i in waste producer group g (%); cx,g,i, coverage of on-site collection of mixed waste for central sorting in waste producer group g including properties from which material i is not separately collected (%); cy,g,i, coverage of on-site collection of mixed waste for central sorting in waste producer group g including properties from which material i is separately collected as on-site collection (%); cz,g,i, coverage of on-site collection of mixed waste for central sorting including properties from which material i is separately collected as drop-off centre collection (%); ed,g,i, separation efficiency of material i in drop-off centre collection in waste producer group g (wt.%); ei, separation efficiency of material i in central sorting plant (wt.%); eo,g,i, separation efficiency of material i in on-site collection in waste producer group g (wt.%); pd,g,i, participation rate of material i in drop-off centre collection in waste producer group g (%); po,g,i, participation rate of material i in on-site collection in waste producer group g (%); R, total recovery rate (wt.%); Rg, recovery rate of waste producer group g (wt.%); Rg,i, recovery rate of material i in waste producer group g (wt.%); r1,g,i, recovery rate of material i in waste producer group g which is achieved with

J.-H. Tanskanen / Resources, Conser6ation and Recycling 30 (2000) 111–133

115

source separation (wt.%); r2,g,i, recovery rate of material i in waste producer group g which is achieved with central sorting (wt.%); sg, the share of waste produced by waste producer group g in total waste (%); sg,i, the share of material i in waste amount produced by waste producer group g (%). Thirdly, the sizes of waste streams in the waste management system and the accumulations of waste types (mixed waste and recoverable materials) at the average property and drop-off centre of each waste producer group are calculated. Waste streams and waste types are described by their waste components. Thus, the effect of separation on the composition of, for example, mixed waste is calculated by the model. Fourthly, collection systems are planned separately for each waste type, waste producer group and separation strategy. The types and numbers of bins and containers and collection frequencies are dimensioned on the basis of the accumulations of waste types at the average collection points. Fifthly, the unit costs and unit emissions of functional elements are determined. The unit costs are connected to the sizes of waste streams. The unit emissions are determined separately for each waste component of a waste stream and expressed, for example, as kg CH4 t − 1 of biowaste landfilled. Sixthly, costs and emissions of MSWM are calculated as a product of the sizes of waste streams and the unit costs and unit emissions (Eqs. (8) and (9)). Finally, the costs and emissions of MSWM can be minimized by iteration, i.e. by replanning separation strategies and collection systems. T =%f %g %i (uf,g,i ×mf,g,i )

(8)

Oc =%f %g %i %j (hc, f,g,i, j ×mf,g,i, j )

(9)

where hc, f,g,i, j is the unit emission c of functional element f resulting from treatment of waste component j which is a part of waste type i in waste producer group g (e.g. mg CH4 t − 1); mf,g,i, amount of waste type i in waste producer group g which is treated with functional element f (t year − 1); mf,g,i, j, amount of waste component j in waste type i produced by waste producer group g and treated with functional element f (t year − 1); Oc, total amount of emission component c (e.g. t CH4 year − 1); T, costs of MSWM (EUR year − 1); uf,g,i, unit cost of functional element f for waste type i produced by waste producer group g (EUR t − 1). The modelling concept described above is based on an analysis of the coverages of on-site collection systems and corresponding accumulations of waste materials at the properties. The analysis of the coverages is based on the fact that large properties are usually obliged to participate in on-site collection of recoverable materials before smaller ones. Thus, the coverages of on-site collection systems can be determined on the basis of the size distribution of properties. In Finland, the minimum size of a property obliged to participate in on-site collection of a material, termed on-site obligation limit, is determined on the basis of the number of households in residential properties and on the basis of the amount of a material produced in commercial establishments. For example in the Helsinki study, 50 185

116

J.-H. Tanskanen / Resources, Conser6ation and Recycling 30 (2000) 111–133

residential properties and 17 597 commercial establishments were included in the analysis in which the size distributions of these properties were calculated. The modelling concept developed can be applied to all regions, municipalities and districts provided that: “ The properties from which source-separated materials are collected on-site are selected on the basis of their size, e.g. the number of households. “ Adequate input data are available.

2.2. The HMA model The HMA model was developed for integrated analysis of separation strategies and their effects on recovery rates, costs and emissions of MSWM (Fig. 2, Table 1). Waste producers were divided into three groups: (1) residential properties smaller than five households (detached houses and small terraced houses), (2) residential properties larger than or equal to five households (terraced houses and apartment houses) and (3) commercial establishments. In addition to mixed waste, source separation of seven materials was included in the model: paper, cardboard, biowaste, energy waste, glass, metal and liquid packaging board, e.g. juice cartons. Energy waste may consist of paper, cardboard, plastics, liquid packaging board and miscellaneous combustible waste components. These combustible waste components can be also sorted and processed centrally for energy recovery. The collection systems of source-separated materials include both on-site collection and drop-off centres, which are defined on the basis of coverage, participation rate and separation efficiency. The HMA model is a static and linear simulation model in the format of an Excel spreadsheet (version 5.0). Nine emission components from collection, backyard composting, central composting and landfilling were included in the HMA model (Table 1). The individual emission components were expressed as four groups of emissions as follows: 1. nutrient load (O2 consumption) consisting of COD, NOx, NH4 and NH3; 2. greenhouse gas load (CO2 equivalents) consisting of CO2, CH4 and N2O; 3. acid load (SO2 equivalents) consisting of SO2, NOx and NH3; 4. ozone formation (C2H4 equivalents) consisting of VOCs. The coefficients needed to convert the individual emission components to the equivalents of emission groups were selected to correspond to the Scandinavian environmental conditions by Pelkonen et al. [28] from the data compiled by the Nordic Council of Ministers [29].

2.3. Study area and input data The Helsinki Metropolitan Area consists of four municipalities, i.e. Espoo, Helsinki, Kauniainen and Vantaa, covering a total of 764 km2. The number of inhabitants in the region was 891 000 in 1995 and the amount of municipal solid waste produced 520 000 t (585 kg person − 1 year − 1).

J.-H. Tanskanen / Resources, Conser6ation and Recycling 30 (2000) 111–133

117

Fig. 2. Graphic presentation of the HMA model (C, bins and containers; CO, collection; Tr, transportation; Ts, transfer station).

In 1995, five types of materials were collected separately in the Helsinki region. Paper was collected on-site from residential properties bigger than or equal to five households and paper and cardboard from commercial establishments in which the production of these materials was more than 50 kg week − 1. Separate collection of biowaste was carried out in one-quarter of the region with on-site obligation limits of 10 households and 50 kg week − 1. In addition, there were drop-off centres for

118

J.-H. Tanskanen / Resources, Conser6ation and Recycling 30 (2000) 111–133

paper, cardboard, glass and liquid packaging board. Waste types were collected in 0.12 – 0.6 m3 bins or with 1.3 – 6.0 m3 containers and compacting collection vehicles. There was one transfer station, one composting plant and one landfill in the region. Nine percent of residential properties composted their biowaste on the spot. The calculation bases of the input data used in the Helsinki study have been presented in details by Tanskanen [26] and by Pelkonen et al. [28]. The input data can be divided into the following three groups: 1. The data needed to calculate the waste amounts and recovery rates (Table 2). These data were based on unpublished statistics compiled by the Helsinki Metropolitan Area Council and by Statistics Finland. In addition, an earlier study of waste composition in the Helsinki region was utilized [30]. 2. The unit costs of the functional elements (Tables 3 and 4). The unit costs were mainly calculated on the basis of the charges levied in the Helsinki region in 1995 and the empirical cost functions supplied by the Helsinki Metropolitan Area Council. Both fixed and operational costs were included in the calculations. The unit costs of waste collection were updated between the strategies studied (Table 8) on the basis of the changes in the accumulations of waste types at the collection points. 3. The unit emissions of the functional elements (Tables 5 and 6). The coefficients used to convert the unit fuel consumption to unit emissions were the following [28]: 154.5 g O2 l − 1 for nutrient load, 2.7 g CO2 l − 1 for greenhouse gas load, 18.1 g SO2 l − 1 for acid load and 2.7 g C2H4 l − 1 for ozone formation. Table 1 Functional elements, costs and emission components of MSWM included in the HMA model Functional element

Costs

Emission components

Waste collection Bins and containers at the properties Containers at drop-off centres Structures of collection points Collection work at the collection area Transportation Transfer station Backyard composting Central composting

Yes Yes Yes Yes Yes Yes Yes Yes Yes

– – – CO2, NOx, SO2, VOCs CO2, NOx, SO2, VOCs – CO2, CH4, N2O, NH3, VOCs COD, CO2, CH4, N2O, NH3, NOxa, NH4, SO2a, VOCs –

Yes

–

Processing of source separated energy waste Central sorting and processing of mixed waste Landfilling Decomposition of waste Landfill compactors Recovery of landfill gas Waste tax Re6enues from reco6ered materials a

Yes

Yes Yes

COD, NH4, CO2, CH4, VOCs CO2, NOx, SO2, VOCs CO2, NOx, SO2,VOCs – –

Emissions from the production of energy needed in composting.

J.-H. Tanskanen / Resources, Conser6ation and Recycling 30 (2000) 111–133

119

Table 2 Data used in the calculation of waste amounts and recovery rates Parameter

Number of inhabitants Number of employees Waste generation rate (t person−1 year−1) Waste composition (wt.%) Paper Cardboard Biowaste Glass Metal Plastics Liquid packaging board Textiles Miscellaneous combustibles Miscellaneous non-combustibles Co6erages of on-site collection systems and corresponding accumulations of materials Current separation acti6ities (wt.%) On-site collection Drop-off centre collection Target separation acti6ities (wt.%) On-site collection Drop-off centre collection Separation efficiency of the central sorting plant (wt.%)

Residential properties 891 000 – 0.265 29 1 28 4 4 7 2 5 16 4 See Fig. 3

Commercial establishments – 412 000 0.690 20 17 30 2 3 7 – 2 16 3 See Fig. 4

50–75 20–50

50–75 –

60–90 50 90

70–90 – 90

3. Results

3.1. Formulation of separation strategies A total recovery rate of 27wt.% was attained with the separation strategy used in the Helsinki region in 1995 (termed Strategy I). The analysis done proved that high coverages were reached in the on-site collection of paper and cardboard with the present on-site obligation limits of five households and 50 kg week − 1 (Table 7, Figs. 3 and 4). However, three major weak points were identified in the separation strategy applied. Firstly, energy waste and metal were not separately collected. Secondly, separate collection of biowaste was applied in only one quarter of the region, resulting in coverages of 19% for residential properties and of 23% for commercial establishments with the present on-site obligation limits of 10 households and of 50 kg week − 1. Thirdly, the present separation activities, 20–75 %, were far from the estimates of the highest achievable activities, 50–90%. Following the analysis above, two different separation strategies were formulated and studied with the HMA model (Table 8). The waste management system used in 1995 (Strategy I) was studied to serve as the point of comparison. Strategy II was based on source separation only and it was formulated by amending the present strategy in the following phases:

J.-H. Tanskanen / Resources, Conser6ation and Recycling 30 (2000) 111–133

120 “

“

“

“ “

Separate collection of biowaste was extended to cover the whole region of Helsinki with the current on-site obligation limits. As a result, the total recovery rate rose from the present 27 to 36wt.%. Separate collection of energy waste started, increasing the total recovery rate from 36 to 47wt.%. The on-site obligation limits used in this calculation were five households and 50 kg week − 1. Separate collection of metal and glass from commercial establishments were started with an on-site obligation limit of 50 kg week − 1. Also, drop-off centre collection of metal was started. The total recovery rate increased from 47 to 48wt.%. The present separation activities were replaced by the target activities, resulting in a total recovery rate of 60wt.%. Finally, the on-site collection systems of paper, biowaste and energy waste were extended to cover all residential properties and the on-site obligation limits of all

Table 3 Types of bins and containers and unit costs of waste collection in the strategies studied Waste type

Type of bin or container (m3)

Residential properties (]5 households Mixed waste 0.60 Paper 0.60 Biowaste 0.24 Energy waste 0.60 Residential properties (B5 households) Mixed waste 0.15 Paper 0.12 Biowaste 0.12 Energy waste 0.12 Commercial establishments Mixed waste 0.60 Paper 0.60 Biowaste 0.24 Energy waste 6.00 Cardboard 6.00 Glass 0.60 Metal 0.60 Drop-off centres Paper 4.00 Cardboard 4.00 Glass 1.30 Metal 4.00 Liquid packaging board 6.00

Costs of waste collection (EUR t−1)

Strategy I

Strategy II

Strategy III

66.3 48.6 154.4 –

71.3 46.8 155.2 117.7

71.0 46.8 149.3 117.7

98.7 – – –

131.2 210.9 742.5 402.6

103.6 – – –

59.0 30.9 102.3 – 122.9 – –

65.1 31.8 104.3 132.9 133.0 143.6 74.0

64.1 30.4 100.9 122.3 119.2 125.0 67.3

42.4 223.5 74.7 – 58.2

– – 74.7 44.4 –

42.4 223.5 74.7 44.4 58.2

J.-H. Tanskanen / Resources, Conser6ation and Recycling 30 (2000) 111–133

121

Table 4 Unit costs of functional elements and revenues from recovered materials Functional element Transfer of mixed waste Transfer of glass Transport of mixed waste from transfer station to landfill Transport of glass from transfer station to markets Transport of energy waste from processing to markets Backyard composting Residential properties]5 households Residential propertiesB5 households Central composting Processing of source-separated energy waste Central sorting and processing of mixed waste Waste tax for final disposal Landfilling Strategy I (380 000 t year−1) Strategy II (180 000 t year−1) Strategy III (135 000 t year−1) Re6enues from reco6ered materials Paper Cardboard Biowaste Energy waste Glass Metal Liquid packaging board

Unit cost (EUR t−1) 10.9 1.7 5.2 13.5 8.4 95.0 645.0 42.0 33.6 33.6 15.1 7.9 10.9 14.3 44.4 42.0 1.7 20.2 8.4 0 25.2

materials were reduced from 50 to 20 kg week − 1. As a result, the total recovery rate increased from 60 to 66wt.%. In Strategy III, Strategy II was complemented with central sorting of mixed waste, resulting in a total recovery rate of 74wt.%. At the same time, Strategy II was modified by stopping separate collection of paper and energy waste from residential properties smaller than five households and separate collection of biowaste from properties smaller than 10 households. For commercial establishments the on-site obligation limits of all materials were raised from 20 to 50 kg week − 1. Also, drop-off centre collection of paper, cardboard and liquid packaging board was started. Strategies II and III represent different, partly alternative and partly complementary choices to aim at the Finnish recovery rate targets of 50wt.% (in 2000) and 70wt.% (in 2005). Strategy II can be regarded as an ultimate strategic goal in source separation in the Helsinki region. For this reason it was selected for further analysis, despite the fact that the recovery rate attained was four percentage units below the target of 70wt.%.

122

J.-H. Tanskanen / Resources, Conser6ation and Recycling 30 (2000) 111–133

3.2. Costs and waste streams of MSWM The costs of MSWM increased by 41% in Strategy II and by 30% in Strategy III compared to the year 1995 (Fig. 5). In Strategy II, residential properties smaller than five households caused 45% of the increase in the total costs. However, the share of these properties of the increase in the total recovery rate was only 10%. In Strategy III, the increase in the costs of MSWM was mainly caused by residential properties bigger than or equal to five households and by commercial establishments. In 1995, the costs of MSWM were 41 400 000 EUR in the Helsinki region (79.3 EUR t − 1 waste and 46.5 EUR inhabitant − 1). The most important functional element increasing the costs of MSWM was waste collection (Table 9). The increase in the costs of waste collection was smaller in Strategy III than in Strategy II because in Strategy III recoverable materials were no longer collected separately from residential properties smaller than five households. Central sorting of mixed waste was an important functional element increasing the total costs in Strategy III. Processing of source-separated energy waste and central composting increased the total costs both in Strategy II and in Strategy III, Table 5 Unit fuel consumption of waste collection in the strategies studied Waste type

Unit fuel consumption (l t−1) Strategy I

Residential properties (]5 households) Mixed waste Paper Biowaste Energy waste Residential properties (B5 households) Mixed waste Paper Biowaste Energy waste Commercial establishments Mixed waste Paper Biowaste Energy waste Cardboard Glass Metal Drop-off centres Paper Cardboard Glass Metal Liquid packaging board

Strategy II

Strategy III

8.8 10.7 17.8 –

12.3 9.2 17.8 16.5

12.3 9.2 15.7 16.5

21.0 – – –

24.4 20.6 122.9 56.1

22.2 – – –

8.9 5.4 9.3 – 9.9 – –

14.0 5.9 10.0 6.3 11.2 10.1 9.3

12.6 4.9 8.8 5.6 9.5 7.5 6.6

5.2 25.5 11.2 – 6.9

– – 11.2 6.0 –

5.2 25.5 11.2 6.0 6.9

J.-H. Tanskanen / Resources, Conser6ation and Recycling 30 (2000) 111–133

123

Table 6 Unit emissions of backyard composting, central composting and landfilling Functional element Unit emissions Nutrient load (kg O2 t−1) Backyard composting Biowaste 25.6 Central composting (windrow) Biowaste 2.2 Landfilling b Paper 3.7 Cardboard 4.4 Biowaste 9.9 Glass 0 Metal 0 Plastics 0 Liquid packaging 0 board Textiles 24.5 Miscellaneous 2.9 combustibles Miscellaneous 0 non-combustibles

Greenhouse gas load (kg CO2 t−1)

Acid loada(kg SO2 t−1)

Ozone formation (kg C2H4 t−1)

83.9

3.01

0.10

15.8

0.05

0.10

246.4 296.8 164.7 0 0 3.3 3.3

−0.21 −0.26 −0.13 0 0 0.02 0.02

0.10 0.11 0.06 0 0 0 0

214.5 221.7

−0.18 −0.19

0.09 0.08

3.3

0.02

0

a The negative values result from the utilization of landfill gas in energy production to replace fossil fuels. b The unit emissions from landfilling were limited to cover 15 years after disposal.

because of the greater amount of waste treated. The costs caused by landfilling and by the governmental waste tax decreased because of the reduced amount of waste disposed of to the landfill. The revenues from recovered materials also increased. The costs of waste collection increased from Strategy I to Strategy II and to Strategy III, although the amount of waste collected did not change. This was due to the following two reasons: (1) separate collection of new types of materials (energy waste and metal) and (2) extended on-site collection of materials. These measures divided mixed waste into several separate waste streams at the properties, resulting in reduced amount of waste collected per pickup. Consequently, the pickup times increased and the efficiency of waste collection was reduced (Fig. 6). Pickup time is the time used at the collection area per tonne of waste collected. The amount of waste collected per pickup affects the pickup time because the time used for preparations before loading at a property and the driving time between properties do not depend on the amount of waste collected.

J.-H. Tanskanen / Resources, Conser6ation and Recycling 30 (2000) 111–133

124

The separation strategies studied increased the recovery rates of all waste materials and affected the waste streams in the Helsinki region (Table 10). The amount of waste directly disposed of to the landfill was reduced from 380 000 t year − 1 in Strategy I to 180 000 t in Strategy II and to 135 000 t in Strategy III. The composition of waste disposed of to the landfill, for example in Strategy II, was as follows: biowaste 35wt.%, miscellaneous combustible waste 15wt.%, textiles 10wt.%, miscellaneous non-combustible waste 10wt.%, paper 9wt.%, plastics 6wt.%, cardboard 5wt.%, metal 5wt.%, glass 4wt.% and liquid packaging board 1wt.%. Table 7 Analysis of the separation strategy used in the Helsinki region in 1995 (Strategy I) Material

Coverage of the collection system (%) Commercial establishmentsa

Residential properties On-site collection Paper Biowaste Cardboard Glass Liquid packaging board Energy waste Metal a

Drop-off centre collection

82 19 – – –

18 – 100 100 100

89 23 87 – –

– –

– –

– –

Only on-site collection is applied.

Fig. 3. Coverage of on-site collection of waste materials and average amount of waste produced per property in the residential properties larger than or equal to the on-site obligation limit in the Helsinki region.

Paper Biowaste Cardboard Energy waste Glass Metal Paper Biowaste Cardboard Energy waste Glass Metal Liquid packaging board

II

IIIc 5 10 – 5 – – –

1 1 – 1 – –

5 10 – – –

Residential properties (no. of households)

On-site obligation limit

50 50 50 50 50 50 –

20 20 20 20 20 20

50 50 50 – –

Commercial establishments (kg per week)

b

The upper limit is used for on-site collection and the lower limit for drop-off centre collection. On-site obligation limits were only applied in one quarter of the region. c In addition to source separation, mixed waste is sorted centrally for energy recovery.

a

Paper Biowasteb Cardboard Glass Liquid packaging board

Material

Description of separation strategy

I

Strategy

Table 8 Separation strategies studied with the HMA model in the Helsinki region

Yes – Yes – Yes Yes Yes

– – – – Yes Yes

Yes – Yes Yes Yes

Drop-off centre collection

50–90 60 50–90 70 50–70 50–70 50

90 60 90 70 50–70 50–70

50–75 50 20–75 40 20

Separation activitya

J.-H. Tanskanen / Resources, Conser6ation and Recycling 30 (2000) 111–133 125

126

J.-H. Tanskanen / Resources, Conser6ation and Recycling 30 (2000) 111–133

Fig. 4. Coverage of on-site collection of paper and the average amount of paper produced per property in the commercial establishments larger than or equal to the on-site obligation limit. Paper is shown as an example.

Fig. 5. Share of various waste producer groups of the change of the costs of MSWM in Strategies II and III compared to Strategy I.

3.3. Emissions of MSWM The amount of emissions caused by MSWM reduced from Strategy I to Strategies II and III as follows: nutrient load by 23 and 28%, greenhouse gas load by 37 and 53% and ozone formation by 17 and 33% (Fig. 7). The reason for the reduction in the amount of emissions was the decreased amount of waste disposed of to the landfill. In Strategy I, the total nutrient load was 3100 t year − 1 expressed as O2 consumption, the total greenhouse gas load was 75 300 t year − 1 expressed as CO2 equivalents and the total ozone formation was 36 t year − 1 expressed as C2H4 equivalents.

J.-H. Tanskanen / Resources, Conser6ation and Recycling 30 (2000) 111–133

127

The amount of acid load increased by 125% in Strategy II and by 114% in Strategy III compared to Strategy I (Fig. 7). The reason for the increase was enhanced recovery which reduced the amount of landfill gas available for energy production to replace fossil fuels. The acid load was smaller in Strategy III than in Strategy II because emissions caused by waste collection reduced from Strategy II to Strategy III. In Strategy I, the amount of acid load caused by MSWM was 46 t year − 1 expressed as SO2 equivalents. The emissions caused by waste collection increased by 30% from Strategy I to Strategy II and by 16% from Strategy I to Strategy III because of increased pickup times. The changes in the amount of emissions caused by composting were simply Table 9 Share of various functional elements in the change of the costs of MSWM from Strategy I to Strategies II and III Functional element

Change of total costs (%) Strategy II

Strategy III

Waste collection Central sorting and processing of mixed waste Processing of source separated energy waste Central composting Backyard composting Landfilling Waste tax Revenues from recovered materials

+45 0 +7 +6 0 −2 −7 −8

+22 +17 +6 +6 0 −2 −9 −10

Total

+41

+30

Fig. 6. Interdependence between the amount of waste collected per pickup (expressed as the number of bins) and the pickup time. Separate collection of paper from residential properties larger than or equal to five households is shown as an example. Unit times used for compacting collection vehicle were based on data compiled by the Association of Finnish Civil Engineers [31].

128

J.-H. Tanskanen / Resources, Conser6ation and Recycling 30 (2000) 111–133

Table 10 Recovery rates of waste materials in the strategies studied Material

Recovery rate (wt.%) Strategy I

Strategy II

Strategy III

Paper Cardboard Energy waste Liquid packaging board Biowaste Metal Glass

69 64 0 20 12 0 27

87 83 68 70 58 51 49

98 98 96 97 51 46 44

Total

27

66

74

Fig. 7. Effect of various functional elements on the change of the total emissions of MSWM in Strategies II and III compared to Strategy I.

due to changes in the amount of biowaste treated. In this study, the emissions from the landfill were limited to cover 15 years after disposal and the emissions occurring after this period were ignored. This limit was applied because emissions from collection and composting are generated with much shorter delays than emissions from landfills and it is difficult to compare present and future emissions. However, the effects of unlimited decomposition time on the amount of emissions was studied in the sensitivity analysis (Table 11).

3.4. Uncertainty and sensiti6ity analysis

“ “ “

The potential sources of errors in the Helsinki study are the following: the system boundary of the HMA model, the linearity of the HMA model, the input data used in the study.

–

−0.4 90.5

98.6 93.7

– 93.3

– 92.1 –

–

−1.6 –

– – –

–

–

–

–

−1.2

–

+1.6

−12.1

90.3

90.5

−10.7

–

–

–

+2.3

−2.3

–

–

Strategy III

–

–

–

–

–

–

−9.6

–

–

–

Strategy II

Change of amount of greenhouse gas load

−4.5

Strategy III

Strategy II

Strategy II

Strategy III

Change of costs of MSWM

Total recovery rate

Effect on results of the study (percentage units)

Share of miscellaneous combustible waste −0.9 decreases by 10%-units increasing the share of biowaste in Strategies II and III Share of miscellaneous combustible waste −6.8 decreases by 10%-units, increasing the share of non-recoverable waste in Strategies II and III Separation efficiency of central sorting – increases by 10%-units in Strategy III Separation efficiency of central sorting – decreases by 10%-units in Strategy III Unit costs of collection (EUR t−1 waste) – in the residential properties smaller than five households change by 9 20% in Strategies I, II and III Unit cost of central sorting changes by – 950% in Strategy III Revenues from energy waste change by – 9 50% in Strategies II and III Unit fuel consumption of waste collection – changes by 9 20% in properties smaller than five households Unlimited decomposition time is used for – landfilling in Strategies I, II and III

Change in the input data

Table 11 Effect of changes in the input data on the following results of the study: (1) total recovery rate of 66wt.% in Strategy II and of 74wt.% in Strategy III; (2) increase in the costs of MSWM of 41% from Strategy I to Strategy II and of 30% from Strategy I to Strategy III; and (3) reduction in the amount of greenhouse gas load of 37% from Strategy I to Strategy II and of 53% from Strategy I to Strategy III J.-H. Tanskanen / Resources, Conser6ation and Recycling 30 (2000) 111–133 129

130

J.-H. Tanskanen / Resources, Conser6ation and Recycling 30 (2000) 111–133

The HMA model was planned according to the waste management system used in the Helsinki region in 1995 and to the waste policy of the Helsinki Metropolitan Area Council. Thus, the system boundary of the model is broad enough to study MSWM in the Helsinki region with good reliability. The linearity of the HMA model was taken into consideration by studying MSWM systems as single simulations, for which the unit costs and unit emissions were individually calculated according to the characteristics of the strategies studied. Most of the input data used can be modified without significant changes in the relative superiority of Strategies II and III because of the similarity of these strategies. However, the changes concerning combustible waste components, residential properties smaller than five households and central sorting of mixed waste are of major importance. In the sensitivity analysis, the effects of changes in these input data on the total recovery rate and on the change of the costs and greenhouse gas load of MSWM were studied (Table 11). Changes in the share of combustible waste components affect the total recovery rate attained in Strategy III more than the recovery rate attained in Strategy II. However, moderate changes do not markedly affect the recovery rates obtained if the share of non-recoverable waste does not increase. The total recovery rate in Strategy III is not sensitive to changes in the separation efficiency of the central sorting plant. The unit cost of central sorting is of major importance from the perspective of total costs in Strategy III because of the great amount of mixed waste sorted. The greatest change in the amount of emissions occurred when the unit emissions of final disposal were determined on the basis of infinite decomposition time instead of only 15 years. However, the effect of this change on the difference between Strategy II and Strategy III was small because both of these strategies were affected. 4. Discussion and conclusions The case study performed in the Helsinki region demonstrated that the HMA model is a suitable tool for the strategic planning of integrated MSWM. Firstly, the analysis of collection systems helps to identify potential separation strategies and to calculate the amounts of materials collected for recovery. Secondly, modelling of MSWM systems makes it possible to determine the effects of separation strategies on the costs and emissions caused by the whole MSWM. The HMA model differs from most earlier strategic planning models by reason of a method developed to analyse the on-site collection systems used for waste materials separated at source for recovery. As a result, the HMA model has the following major advantages compared to most other models: (1) The coverages of on-site collection systems of materials can be adapted to the characteristics of a study area; (2) The recovery rates and sizes of waste streams can be calculated on the basis of the characteristics of the separation strategies instead of giving them as input data; (3) The unit costs and unit emissions of waste collection can be updated between separation strategies because the changes in the amounts of materials separated at the properties are known.

J.-H. Tanskanen / Resources, Conser6ation and Recycling 30 (2000) 111–133

131

The modelling concept developed can be applied to all regions, municipalities and districts to study the effects of separation on MSWM provided that: “ the properties from which source-separated materials are collected on-site are selected on the basis of their size, e.g. the number of households; “ adequate input data are available. The Helsinki study indicates that the national recovery rate target of 70wt.% adopted for municipal solid waste in Finland can only be achieved by adding central sorting of mixed waste to source separation strategies. A recovery rate of 66wt.% was reached by a source separation strategy in which separate collection of recoverable materials covered all residential properties and 93% of commercial establishments. In addition, the estimates of the highest attainable separation activities were used in calculations. At the same time, the costs of MSWM increased by 41% compared to the present situation. By supplementing source separation with central sorting of mixed waste, a recovery rate of 74wt.% was attained and the increase of total costs was 30% compared to the present situation. The separation strategies studied reduced the nutrient load, greenhouse gas load and ozone formation caused by MSWM. The reason for this was the reduced amount of waste disposed of to the landfill. The acid load increased for the same reason, because less landfill gas was available for energy production to replace fossil fuels than in the present situation. The combination of source separation and central sorting resulted in a smaller amount of emissions than source separation alone. This was because the central sorting reduced both the amount of waste landfilled and collection work. Universal conclusions about the effects of separation of individual waste materials on the costs and emissions of MSWM cannot be drawn on the basis of this study for two reasons. Firstly, the effects of source separation vary depending on several factors, e.g. characteristics of the region in question, the type of materials separated and the collection method applied. Secondly, this study did not cover all emissions caused by MSWM, e.g. emissions caused by burning of energy waste. Acknowledgements This work was carried out at the Finnish Environment Institute. The study was financed by the Helsinki Metropolitan Area Council, a joint municipal organisation with overall responsibility for waste management in the Helsinki region, and by the Ministry of the Environment, Finland. Special thanks are extended to Jukka Paavilainen and Jarmo Nurmivaara of the Helsinki Metropolitan Area Council for valuable information and successful cooperation during the study. I acknowledge Markku Pelkonen and Elisa Rauta of the Helsinki University of Technology for producing most of the unit emissions and the weighting factors. The author also acknowledges Professor Matti Melanen of the Finnish Environment Institute and Juha Kaila Dr. Tech. of the Helsinki Metropolitan Area Council for providing constructive comments that greatly improved the paper. The language of the manuscript was revised by Virginia Mattila.

132

J.-H. Tanskanen / Resources, Conser6ation and Recycling 30 (2000) 111–133

Appendix A. Key definitions Coverage of a collection system: in an area the ratio of (a) the amount of a material produced in those properties where separate collection is available and (b) the amount of the material in question produced in all properties of the area. On-site obligation limit: the minimum size of a property obliged to participate in on-site collection of a material in an area. Participation rate: the share of people providing sorted material to bins in those properties where separate collection is available. Pick-up time: the time used at the collection area per tonne of waste collected. Recovery rate: the share of waste which is separated and delivered to material or energy markets. Separation activity: the share of a material which is correctly separated in those properties where separate collection is available. Separation activity consists of participation rate and separation efficiency. Separation efficiency: the share of a material which is correctly separated by the people who participate in separation. Also, the share of a material which is correctly separated in a central sorting plant. Waste stream: separate waste output of, e.g. a property, functional element or study area. Waste type: mixed waste and recoverable waste materials.

References [1] Tchobanoglous G, Theisen H, Vigil SA. Integrated Solid Waste Management: Engineering Principles and Management Issues. New York: McGraw-Hill, 1993. [2] Gottinger HW. A computational model for solid waste management with application. Eur J Oper Res 1998;35(3):350–64. [3] MacDonald ML. Solid waste management models: A state of the art review. J Solid Waste Technol Manage 1996;23(2):73–83. [4] Rushbrook P. Waste management planning: An illustrated description of ‘HARBINGER’, the Harwell Waste Management Model. Waste Manage 1987;77(6):348– 61. [5] Chapman RE, Berman EB. The resource recovery planning model: A new tool for solid waste management. U.S. Department of Commerce, National Bureau of Standards, NBS Special Publication 657. Washington DC, 1983. [6] Kaila J. Mathematical model for strategy evaluation of municipal solid waste management systems. Technical Research Centre of Finland, Publications 40, Espoo, Finland, 1987. [7] Lund JR. Least-cost scheduling of solid waste recycling. J Environ Eng 1990;116(1):182– 97. [8] Jacobs TL, Everett JW. Optimal scheduling of consecutive landfill operations with recycling. J Environ Eng 1992;18(3):420–9. [9] Zach PA. Life cycle inventory analysis as applied to recycling system economic costs. J Res M T 1992;20(4):181–9. [10] Baetz BW, Neebe AW. A planning model for the development of waste material recycling programmes. J Oper Res Soc 1994;45(12):1374– 84. [11] Anex RP, Lawver RA, Lund JR, Tchobanoglous G. GIGO: Spreadsheet-based simulation for MSW systems. J Environ Eng 1996;122(4):259– 62. [12] Everett JW, Modak AR. Optimal regional scheduling of solid waste systems. I: Model development. J Environ Eng 1996;122(9):785–92.

J.-H. Tanskanen / Resources, Conser6ation and Recycling 30 (2000) 111–133

133

[13] Huhtala A. A post-consumer waste management model for determining optimal levels of recycling and landfilling. Environ Resour Econ 1997;10(3):301– 14. [14] Ansems AMM, Langerak E. Model for the collection and processing of waste streams, In: Sundberg J, Nybrandt T, Sivertun A, , editors. Systems Engineering Models for Waste Management. Proceedings of an International Workshop, 25 – 26 February 1998. Gothenburg, Sweden: Swedish Environmental Protection Agency, 1998, pp. 129 – 134. [15] Haith DA. Material balance for municipal solid-waste management. J Environ Eng 1998;124(1):67– 75. [16] Tanskanen J-H, Melanen M. Modelling separation strategies of municipal solid waste in Finland. Waste Manage Res 1999;17(2):80–92. [17] Pictet J, Giovannoni J-M, Maystre LY. Impact assessment of urban waste processing systems using a multibox model. Water Air Soil Pollut 1992;63(1-2):155– 78. [18] Sundberg J. Generic Modelling of Integrated Material Flows and Energy Systems. Gothenburg, Sweden: Chalmers University of Technology, Department of Energy Conversion, 1993. [19] White PR, Franke M, Hindle P. Integrated Solid Waste Management: A Life Cycle Inventory. Glasgow: Blackie, 1995. [20] Ljunggren M. A Systems Engineering Approach to National Solid Waste Management. Gothenburg, Sweden: Chalmers University of Technology, Energy Systems Technology Division, 1997. [21] Tanskanen J-H, Reinikainen A, Melanen M. Waste streams, costs and emissions in municipal solid waste management: A case study from Finland. Waste Manage Res 1998;16(6):503– 13. [22] Wang FS, Richardson AJ, Roddick FA. SWIM — Modelling approach and its applications, In: Sundberg J, Nybrandt T, Sivertun A, , editors. Systems Engineering Models for Waste Management. Proceedings of an International Workshop, 25 – 26 February 1998. Gothenburg, Sweden: Swedish Environmental Protection Agency, 1998, pp. 121 – 127. [23] Kaila J. Service quality factors in waste collection with increased source segregation and producer responsibility. 7th ISWA International Congress, 27 October – 1 November 1996, Yokohama, Japan, Vol. 2, pp. 188–189. [24] Gielen DJ. The MARKAL systems engineering model for waste management, In: Sundberg J, Nybrandt T, Sivertun A, , editors. Systems Engineering Models for Waste Management. Proceedings of an International Workshop, 25–26 February 1998. Gothenburg, Sweden: Swedish Environmental Protection Agency, 1998, pp. 31– 51. [25] Ministry of the Environment, Finland. Valtakunnallinen ja¨tesuunnitelma vuoteen 2005 (The national waste plan until 2005). The Finnish Environment 260. Edita Ldt, Helsinki, Finland, 1998 (in Finnish). [26] Tanskanen J-H. YTV:n alueen ja¨tehuollon mallintaminen (Modelling municipal solid waste management in the Helsinki Metropolitan Area). YTV Helsinki Metropolitan Area Council, Helsinki Metropolitan Area Series C 1997:2, Helsinki, Finland, 1997 (in Finnish). [27] Tanskanen J-H. Modeling source separation strategies in the Helsinki Metropolitan Area, In: Sundberg J, Nybrandt T, Sivertun A, , editors. Systems Engineering Models for Waste Management. Proceedings of an International Workshop, 25 – 26 February 1998. Gothenburg, Sweden: Swedish Environmental Protection Agency, 1998, pp. 165 – 175. [28] Pelkonen M, Rauta E, Tanskanen J-H. Yhdyskuntaja¨tehuollon yksikko¨pa¨a¨sto¨t YTV:n alueella (Unit emissions of municipal waste management in the Helsinki Metropolitan Area). Unpublished manuscript, in Finnish. Helsinki, Finland: Helsinki University of Technology, 1996. [29] Nordic Council of Ministers. Nordic guidelines on life-cycle assessment. Nord 1995:20, Copenhagen, Denmark. [30] Helsinki Metropolitan Area Council. Pa¨a¨kaupunkiseudun yhdyskuntaja¨tteen koostumus 1990 (The composition of municipal solid wastes at the Helsinki Metropolitan Area in 1990). Helsinki Metropolitan Area Series C 1991:3, Helsinki, Finland (in Finnish). [31] Association of Finnish Civil Engineers. Ja¨tehuolto: Yhdyskuntaja¨tteen kera¨ily ja kuljetus (Waste management: Collection and transportation of municipal solid waste). RIL 152-1983, Helsinki, Finland (in Finnish), 1983.