Automated and manual collection systems each have strengths and weaknesses. Automated collection can provide a high service level to residents at low cost but it requires a major process change and capital commitment. It’s relatively inflexible for service adjustment and takes away some collection control. Manual collection demands more from residents and is a more costly service; however, it gives control to collectors and is inherently flexible.
There is no “correct” answer in the automated versus manual debate. Each community must determine its waste management goals and choose a suitable collection system to attain those goals within various constraints.
In 1985, the City of Regina, Saskatchewan implemented automated collection for part of its residential waste stream. This system has been primarily applied to houses with back-alley access while houses with front-street-only access have mostly stayed on a manual collection system.
Regina is a medium-size western city (population 190,000). Annual temperature extremes range from minus 40 to plus 40 degrees Celsius. Precipitation is low at 360mm per year (half of which is snow) and there are strong winds. The city is flat and compact without any major geographical features such as rivers, hills or valleys. It covers approximately 120 square kilometres, has 900 km of streets and 300 km of back-alleys.
All waste is direct hauled to the municipal landfill.
The initial plan was to completely convert residential collection from manual to automated in two stages over several years. The first stage involved the conversion of 27,000 houses with back-alley access to shared container automated collection using large steel containers in the rear lanes. A further 4,000 front-street houses were included. This stage was completed in 1988.
The second stage was to be the conversion of all other houses (essentially those which did not have access to rear lanes) to automated collection using plastic carts which residents would roll out to the street on collection day. The city decided not to implement the second stage for cost and operational reasons.
Manual collection equipment consists of a fleet of side-loading trucks with tandem axle chassis and 22-cubic-metre bodies. A two-person crew is assigned to each truck with one driving and one collecting. They alternate tasks to equally distribute labour and to provide relief from hot and cold weather extremes.
In 2000 the number of routes in each area increased from five to six due to housing growth. Statutory holidays (11) and corporate earned days off (17) reduce the normal workweek by 28 days per year. Therefore, houses with manual collection average about 47 pickup cycles per year, and there are times (especially around Christmas and Easter) where the interval between pickups may be as long as 12 days. Houses on manual collection receive a calendar that specifically identifies pickup days for a two-year period.
The amount and types of waste are monitored and enforced by collection personnel. Volume is restricted to eight containers or 750 litres per visit. Excess volumes or unacceptable types of waste may be left by collection staff, in which case they “tag” the waste to inform the resident and register the infraction.
Automated collection makes use of a fleet of special robotic trucks and a large number of specialized containers. One person who remains in the truck cab operates the trucks. From the starboard side of the truck, the operator uses a robotic system to hook, lift and tip waste from the container into the truck body.
The 6,000 rear lane containers are made of steel (with plastic lids) and have volumes of 2,250 or 1,125 litres. The containers are provided by CIM of Humboldt, Saskatchewan and by Whittke of Medicine Hat, Alberta. Each resident makes use of 500 litres of volume on average. The 4,000 front-street rollout containers are made of plastic and have a volume of 450 litres. General Body and Equipment of Calgary, Alberta provides the plastic lids.
Automated collection is done on an area basis with a simple weekly cycle that is not changed by statutory holidays or days off. There are seven routes on the first four days of the week and five routes on Friday. Residents may put their garbage in the shared back-alley containers at any time. Only the rollout containers (one per house) require emptying on the same day they are set out. This allows more operational flexibility to deal with equipment breakdown or inclement weather.
The same prohibitions and restrictions for manual collection areas apply to automated collection. However, the nature of automated collection does not allow the collection personnel to decide what to pickup or not. The waste is commingled in the container and the truck operator tips all containers unless it is not possible to do so. (Most “tagging” on the automated system relates to the inability to tip a front-street rollout container because of parked cars, snow ridges or improper orientation of the container.)
Comparisons and considerations
Because of system designs, automated collection is slightly more frequent (52 vs. 47 pickups per year) while manual collection allows more waste (750 litres vs. 500 litres) per cycle). Automated collection services 36 per cent more houses/cycles than manual.
Residents use a call centre to discuss service. Most of the 1,943 calls regarding automated collection have been to report damaged or unsanitary containers, or to complain of the location of a rear lane container. Only a small number of calls were service complaints.
Interestingly, half of the 622 calls regarding manual collection were service complaints (garbage left) and half were information requests. Problems that collectors have with residents are measured with “tags issued.” As previously mentioned, automated collection issues very few tags (844). Manual collectors issued significantly more tags (2,030) to indicate restriction on volumes and waste types.
Based on 1,034,106 manual collection visits per year, collectors penalized residents for improper use of the system 0.2 per cent of the time (2,030 tags). However, complaints were received from the residents on only 0.06 per cent of the visits (622 calls). Based on 1,612,981 automated visits per year, collectors penalized residents for improper use of the system only 0.05 per cent (844 tags) of the time. However, residents called against 0.1 per cent (1,943 calls) of visits.
Initially, automated collection was strongly resisted by the community. There was a high level of satisfaction (greater than 90 per cent) with the existing manual collection practices and a fear of the unknown. Significantly, within two years of the change the level of satisfaction increased and the public embraced the automated system.
Residents reported the following operational benefits:
large volume containers accommodate larger, bulkier wastes;
waste can be put in the container at any time, thereby eliminating the need for storage on the residents property;
actual collection day became unimportant as long as there was space in the container when residents wanted to deposit waste; and,
an inability to enforce restrictions on waste type and amount allowed residents to conveniently dispose of unacceptable waste. (This is clearly an operational problem.)
An interesting side benefit is that most neighbourhoods with rear lane collection have experienced an improvement in cleanliness.
But residents who have always had manual collection provide a different perspective on service. Since the collectors can make a decision on what to take this system is subject to higher operational control and residents are required to follow the rules. In fact, the rules are not particularly onerous and relatively easy to circumvent. At 750 litres per cycle the volume limit is high. (It equals eight garbage cans.) Moreover, if garbage is properly packed it’s difficult to determine if there’s unacceptable waste (except for extremely heavy materials suc
h as concrete and sod). Therefore, residents have used ingenuity to “hide” unacceptable wastes in the garbage and “share” excess volumes of waste with their neighbours who have not put out the maximum.
Measurements of service
Household generation, productivity and cost all contribute to measurement information that quantifies the performance of each collection system. There has always been a higher generation rate associated with automated collection. However, the difference in 1991 was only 9 per cent more from automated households. By 2000 the difference had grown to 37 per cent more from automated households.
Generation rates have decreased for manual collection households and remained more or less constant for automated collection households. It’s noteworthy that during this same period the city implemented a modest depot-based paper-recycling program available to all residents.
Automated collectors have maintained a high and steady productivity — close to 20 tonnes per person per day. Manual collectors have stabilized at about 6.5 tonnes per person per day. The per-worker productivity of the automated system is almost three times that of the manual system.
Table 2 provides the complete operating cost, including overhead burdens and indirect costs. This includes time-loss costs (Workers Compensation) which are virtually identical for the automated and manual systems although it might be expected to be higher for manual collection. The capital cost for the 10,000 automated containers is not included in this analysis because it was completely repaid by 1992. (However, the steel rear lane containers now need repair or replacement because of rust on the container bottoms; the plastic front-street containers are also wearing out and may require replacement in the future.)
The high productivity is reflected in the much lower weight-unit cost of both front street and rear lane automated collection than for manual collection. However, because more weight units are collected per house with the automated system the house-unit costs are much closer. While the cost of rear lane automated collection is still much less than that for manual collection, front-street automated collection is actually slightly more expensive than manual collection.
Derrick Bellows is with the Public Works Department of the City of Regina, Saskatchewan.
Table 1. A “head to head” comparison of the systems for 2000. A “cycle” is one complete pickup cycle.
|1,943 resident calls
||622 resident calls
|844 tags issued
||2,030 tags issued
Table 2. The 2000 costs accurately reflect the complete operating cost, including overhead burdens and indirect costs. This includes time-loss costs (Workers Compensation) which are virtually identical for the automated and manual systems although it might be expected to be higher for manual collection.