An evaluation of cost-effectiveness of source separation models in Italy: tools to integrate and optimise the system

 Favoino E., M. Ricci, A. Tornavacca, R. Ragazzi

INTRODUCTION

The National Law on Waste Management and Disposal (Decree 22/97) states that waste reduction and material recovery must be preferred to energy recovery and landfilling. Disposal is allowed only for non-recyclable or treated materials, and collection must be efficient and economic. The decree specified recycling targets (for each province) of:

• 15 per cent by March 1999
• 25 per cent by March 2001
• 35 per cent by March 2003

In order to achieve these targets, source separation in Italy is undergoing impressive growth. Attention is focusing on the dominant waste fractions (such as paper and compostable organic waste).

Although source separation of organic waste (kitchen and garden waste) is not compulsory, it is becoming the back-bone of the waste management system, yielding (particularly when operated with door-to-door systems) recycling rates as high as 40-60 per cent.

In general, the intensive collection of dry recyclables alone (paper, glass, metal and plastic) will not allow municipalities to recycle rate 35 per cent. Accordingly, most regions and provinces plan to promote food waste source separation from households and major producers (restaurants, canteens, greengrocers, etc.).

By late 1998 (18 months after the Decree was issued), some 600 municipalities were running wet-and-dry collection schemes (see Table 1).

Using these extensive experience, it is possible to assess the effectiveness of these systems (see Table 2) in terms of Quantitative effectiveness – the specific collecting capacity in (g per person day or Kg per person per year) and Purity of the fraction collected.

The survey shows the excellent quality of material collected (typically 97-99 per cent). This compares well with the 95 per cent purity (5 per cent rejects) of high quality composted products which do not require expensive pre-sorting and final refining technologies in composting plants. This situation is the case in Germany and Austria, where the purity of the material collected usually ranges between 93-98 per cent.

The purity of food waste collected in Italy appears related to the specific features of the collection service. Among these features, the use of watertight, transparent bags (usually biodegradable) for the delivery of food scraps is welcomed by householders. This enhances awareness and participation in the programmes.

The watertight nature of the bags accommodates most kitchen residues (including wet or cooked produce, such as meat and fish scraps), reducing the level of putrescible materials in residual waste. It also minimises leaching and odours from bins and buckets. The transparency of the bags helps quality control. Purity usually falls (to 90-95 per cent) were collection systems involve the use of large road containers, instead of a door-to-door service. An evolution of the road container system is used in parts of Emilia-Romagna. Here, each home receives a key to a locked road container. This means asking only the more aware and responsible households. Purity in such cases can be very high (see Nonantola, Table 2).

However, road container systems suffer from receiving much more garden waste than food waste – a big container is an easy tipping site for bulky materials, while the greater distance from households deters delivery of the more difficult, fermentable items, such as food waste.

A major waste management concern in Italy is the cost-competitiveness of source separation systems with high recycling rates, compared to the traditional mixed waste collection.

Analyses usually express the costs per unit mass for a single waste material collected. However, there is evidence that this distorts the true picture, because the more waste is collected, the lower the costs (per kg) of collection appear. This distortion obscure some important aspects of integrated source separation and waste management:

• reduction of total waste delivered
• contribution of home composting programmes to the overall reduction of organic waste collected.

Evaluating a single waste flow does not allow one to compare advantages to collection rounds and costs for other materials, flowing from operational integration. In effect, the collection of food waste allows important changes in the collection scheme, by reducing, for instance, collection frequencies for residual waste (‘restwaste’).

It must be stressed that the cost of the system (collection plus transport) isn’t paid for according to the amount of the waste collected, but in its entirety. It is correct to express the cost not per unit mass, but as cost per person. This permits a fair comparison of the competitiveness of different systems (in terms of cost, quantity and quality of materials recycled).

A FIRST SURVEY WITH THE NEW EVALUATION METHOD: AN OVERVIEW OF COLLECTION COSTS

In order to compare different collection systems, the research group surveyed the costs of three collection systems in:

• traditional source separations, based on the use of plastic bags or road containers (up to 3.3 m³) for mixed MSW, with source separation using road containers for dry recyclables. The food waste is not sorted, but is delivered along with the residual waste - this remains putrescibles and must be collected frequently.
• intensive source separation, based on road containers (< 3.3 m³) both for food waste and dry recyclables; with collection of residual waste also through road containers. This is usually referred to as the ‘double container’ collection.
• intensive source separations with a door-to-door (DtD) collection for food waste and residual waste. Some high-yield dry recyclables can also be collected with a DtD system.

Traditional systems

 Table 3 reports on the costs of such a collection.

The data show that total waste management costs (including disposal) fluctuate widely because of the different disposal fees charged in different regions. Therefore, in order to evaluate the competitiveness and draw reliable conclusions it is necessary to focus on collection and transport costs, disregarding disposal costs.

The results also indicate that data expressed cost per unit mass penalise municipalities production less waste.

The average collection and transport cost of the three municipalities with waste arising below 350 Kg per person per year is ITL253/kg, while municipalities with more than 500 Kg per person per year have costs of ITL134/kg.

In absolute terms, these local authorities dispose of more waste, so overall waste management cost tends to be higher.

The per capita cost of waste management (collection + transport + disposal) averages ITL 112.000 person per year, while the cost of collection + transport (without disposal) ranges from ITL 44.000 - 112.000 per person per year (averaging ITL66.000).

As mentioned earlier, these systems can be grouped into two categories:

• door to door (DtD) collection systems
• road container collection systems

The study focused on established schemes (operating for at least two year), mainly in northern Italy. Table 4 and 5 summarise the service costs (collection + transport + disposal)

The results indicate that collection schemes based on the use of road containers (whether for mixed MSW or separate food waste) show a higher specific waste production than schemes where small waste bins and buckets are given to single households (DtD collection)

The traditional collection systems based on separation of dry recyclables by means of road containers (table 3), surprisingly shows a higher cost per inhabitant than systems with a source segregation of food waste. This is partly due to higher collection frequencies in southern Italy (up to six times a week) that affect the average costs in Table 3.

The most surprising result is that the average collection and transport costs (per person per year) tend to be sharply lower in schemes where source segregation of food waste use DtD systems, than where road containers are used. This contradicts what is generally expected, due to the much higher number of pick-up points in DtD schemes.

Table 5: Cost data for Italian municipalities operating door-to-door source separation for food

One might think that lower costs of the DtD systems are due to the relatively small number of councils examined - this could be important in the evaluation of specific features related to weather conditions (e.g. more frequent collection or bin washing), type of dwellings, etc.. However, costs of different collection systems in the same area were evaluated.

Data from Verona and Venezia indicates that source segregation of food waste with DtD schemes can be run with no substantial increase in overall costs, and sometimes costs are even lower than with traditional collection or with food waste segregation by means of road containers.

Once more, it must be stressed that traditional collection and road containers may seem to be cheaper in terms of cost per unit mass, even though the cost per capita is equal to or higher than DtD systems.

To understand the unexpected survey results, it should be emphasised that if source separation of food waste is merely added to the system, with no modification of the existing scheme, total costs are bound to rise.

This happen with food waste collection in road containers, but not when food collection is integrated using DtD schemes.

Intensive DtD schemes for food waste yield high quantities, lowering the percentage of food waste in residual waste, which is then collected less frequently. Also, food waste needs no compaction – enabling the use of cheaper collection vehicles.

Collection frequencies for residual waste can be cut only when there is effective separation of foodstuff. Table 6 shows that DtD schemes enable much higher performances. Around 170- 240 grams per person per day were reported for food waste.

Large road containers yield much lower quantities (in fact, their capture rate can be similar, but a high percentage of garden waste is present, and food waste recovery is low). We could that collection using road containers results in a lower participation rate.

Cutting collection frequencies for residual waste constitutes one of the most important tools to optimise schemes for food waste source segregation.

It is particularly effective in area where high collection frequencies are usual for MSW collection.

Diversifying the fleet of collection vehicles

Where a DtD schemes for food waste exist using small bags (bins for high-rise buildings and small buckets for single families in houses with gardens), a material with high bulk density (0.5-0.7 kg/litre) is targeted.

This can be collected using non-compacting vehicles, but only when schemes prevent delivery of garden waste along with food waste. Hence, it is advisable to limit the size of containers supplied (6-10 litres for a single family and < 30 litres for group of 3-4 families).

Households can manage garden waste through:

• home composting promoted by the municipality
• delivery to local recycling centres
• DtD garden waste collection with low frequencies (e.g. 1-2 times per month, only in the warm season).

A two-shift scheme for food waste collection using bulk lorries tends to equal the cost of a single-shift collection for residual waste with compactors. With a cost evaluation in cost per kg, the comparison would not favour the collection of food waste, because the quantity collected is lower than that of residual waste (60-80 kg per person per year, versus 100-200 kg per person per year).

However, food waste gets collected at a much lower cost than with traditional mixed collection, so the overall costs of an integrated sorting scheme are similar or lower.

Effective segregation of food waste allows a number of collection shifts (for different waste fractions) that tends to equal the previous schedule (for mixed collection).

For example, one can collect food waste twice weekly and residual waste once per week in northern Italy, while mixed MSW collection is undertaken three times per week. Furthermore, some collection shifts will cost less through the use of low-tech vehicles.

Cost evaluation is further confirmed (see Table 7), when considering numbers reported for Consorzio Cremasco and some municipalities in Bergamo and Verona Provinces.

Again, the cost of twice weekly collection for food waste (using non-compacting vehicles) is comparable to a weekly collection of residual waste with compacting vehicles.

According to the data presented, it is clear that the main mistake made when planning sorting scheme, is the added feature of the scheme. When new collection are run in addition to the previous mixed MSW collection, they cannot yield savings, to fund the new scheme.

It is vital that the new separate collection is integrated into the established waste management system (i.e. changing frequencies and volumes to collect residual waste).

A further tool to optimise the scheme is the use of suitable vehicles to collect food waste, due to its higher bulk density when garden waste is eliminated from the collection scheme.

Smaller municipalities with direct responsibility for MSW collection often own a single collection truck, which is problematic when planning changes.

Nevertheless, higher institutional echelons (the district or province) can help, by buying appropriate vehicles and lend or lease them to small municipalities. This is already happening in two provinces in Central Italy (Chieti and Pescara).

AMIA Verona, "Raccolte differenziate: Banca dati. Indagine 1997", Roma March 1999
Bigliardi, P. "Frazione umida compostabile da utenze domestiche. Esperienze e prospettive", Proc. "Ricicla 98", Rimini September 1998
Cocchi, A. "Risultati delle raccolte differenziate nel Persicetano", personal communication, 1997
Consorzio Bacino Padova Uno, "Indagine sulla raccolta dei rifiuti urbani: anno 1997", Vigonza (PD) May 1998
Consorzio Provinciale della Brianza Milanese, "Rapporto sulla Gestione dei Rifiuti Urbani ed Assimilati: Anno 1997", Seregno (MI) 1998
Degani, P. "La gestione dei R.S.U. nel Bacino VE4: proposte operative", Proc. Conference "Raccolta differenziata e …dintorni", Mira (VE) October 1998
Favoino, E. "Le raccolte differenziate delle frazioni compostabili in ambito urbano", Proc. of the 4^th National Conference "Produzione ed impiego del compost di qualità", Bari, May 1999
Istituto di Ricerche Ambiente Italia, "Effetti economici ed occupazionali della gestione dei rifiuti" Report issued by Federazione Nazionale delle Liste Verdi, Roma September 1997
Kramer, L. Onida, M. (published by) "I rifiuti nel XXI secolo: il caso Italia tra Europa e Mediterraneo", Edizioni Ambiente, Milano May 1999
Legambiente, "Comune Riciclone 1998: Premio ai migliori risultati di riciclaggio dei rifiuti", Roma July 1998
Osservatorio Rifiuti Provincia di Lecco, "Produzione dei Rifiuti Urbani ed andamento delle Raccolte differenziate: rapporto anno 1997", Lecco July 1998
Osservatorio Rifiuti Provincia di Milano, "RSU Elaborazione dati 1997", Milano October 1998
Provincia di Milano in "Il quaderno. Gestione Rifiuti Solidi Urbani 1998, indirizzi programmatici e azioni di approfondimento", Milano December 1998
Ricci, M."Evoluzione delle raccolte differenziate in Provincia di Verona: le esperienze di raccolta secco umido", Proc. Conference "Dalla raccolta Differenziata alla tariffa sui RSU" Sommacampagna (VR) April 1999