A designer from the solid waste business unit sits down at a computer terminal displaying an overview map of the city. With built-in zoom capabilities, a click of the mouse causes the view to change and focus on a selected area. With geographic information systems (GIS) behind it, the map area is far more than just a neat picture.
GIS is a “living database” of information that the designer uses to create different scenarios to answer a variety of questions. For example, with just a few keystrokes and mouse movements, the number of garbage collection vehicles needed to efficiently service an area quickly can be determined.
Vehicle types can be changed on screen and the impacts automatically assessed. Individual routes, or “beats,” within the area can be created from start to finish in a matter of minutes. Planning that used to take months now takes days.
These capabilities and many more now are a reality in a system aptly named Beat 2000, developed in the Solid Waste Services Business Unit of the city of Calgary in Alberta, Canada. “Beat 2000 has become an integral part of the way we do our business,” says Dave Griffiths, general manager of Solid Waste Services. “There is no going back.”
To the Drawing Board
Calgary is a rapidly growing city located in the foothills of Alberta's Rocky Mountains. The city currently has 860,749 residents living in 340,257 dwellings. The city's growth has been explosive: It grew by 23,054 people (or 10,283 dwellings) from 1998 to 1999 and by another 18,361 people (9,098 dwellings) from 1999 to 2000.
Maintaining high-quality refuse collection was becoming impossible using the old, partially computerized, manual mapping system. For example, it took the staff almost one year to create a new schedule (called the base design) just for the winter period.
The summer period was even more complicated. Extra trucks and crewmembers are needed in the summer to collect the increased amounts of grass clippings, leaves and other yard waste. To meet the increased demands of this period (called the peak design), city blocks had to be subtracted from each winter design beat. The remaining blocks then had to be assigned to the seasonal vehicle and crew. (Because of the increase in the amount of garbage per unit, the service area in the summer is less for each truck.)
A summer crew consequently had a beat made up of blocks from one or more base design beats. This means that peak design beats were less efficient.
Using Calgary's old system, both major and minor changes to designs were difficult and labor-intensive. Over time, the colored dots and tapes used on plastic sheets overlaying the manual maps tended to fall off. This lost data could lead to flawed designs.
By the mid 1990s, the city recognized its old system was inefficient and concluded it needed a better alternative. Consequently, over the next several months, the city's solid waste staff evaluated the software, hardware and business practices used in several solid waste operations. This evaluation included site visits. In the end, however, the city decided to develop its own GIS based mapping system.
By the beginning of 2000, a project team designed, developed and built the city's new mapping system on a Vision*/Oracle GIS platform running on a Sun Solaris Unix operating system. The product, the GIS-based Beat 2000, includes three components — an Intelligent Block Face (IBF) polygon and the data within; time studies based on truck type, crew size and haul distances; and the design tools to bring them together.
Capturing the Data
All GIS systems consist of graphics, which act as a data repository, and associated data, which is the specific information that needs to be stored. One of the problems inherent in GIS systems is the relative quality and timeliness of this data. Fortunately, the city of Calgary keeps this information current by conducting an annual civic census and having a property mapping section that maintains an electronic graphic version of all land parcels in the city.
These parcels are imported as graphics into the GIS system and, through tools developed by the team, are used to create a unique feature called an IBF polygon — which is the foundation of Beat 2000. A polygon is a series of connecting lines that enclose an area such as a city block.
A typical GIS system uses a regular polygon, which is a single line of many segments surrounding and related to a center point. Picture a person standing on a hill in the center of one of these polygons who would be able to see all of the structures and lots within the enclosure. A high fence would conceal everything beyond. The viewer would not be able to see what was on the other side, whether it was a street, alley or the back of another lot.
An IBF polygon is multiple lines of multiple segments surrounding and related to a center point. Now, that same person standing on a hill in the center would see all the same structures and lots, but the fence would be very different. The fence bordering an alley or street could be recognized easily by a designated color and other features. In addition, one of these fences would be designated as active. This active fence line would then be treated as the polygon-to-network connection, which allows the system to quickly perform a number of tasks, such as automatically finding the haul distances to the three city-owned landfill sites.
This “intelligent connection” satisfies a direct business need of the city, because in Calgary, 30 percent of residential refuse is collected from the front street and 70 percent from lanes.
Not only is haul distance data stored in these IBF polygons (the city is currently covered in 47,481 polygons), but all lots and lot sizes within the polygons are captured, as are addresses and address changes. All data from the annual civic census, the Solid Waste Services mini-census and other data, such as information collected from commercial bulk customers, also are captured and stored in this first component of Beat 2000.
On Time and By Design
Starting in the early 1970s, the Solid Waste Services Unit conducted time studies ultimately to determine collection design. Today, these studies involve specific truck types, crew sizes, structure types (e.g., single family, duplexes, apartments, rowhouses/condos, basement suites, mobile homes and other structures) and haul distances to the closest landfill. For the summer design, large and small lot sizes (reflecting increased summer yard waste) also are factored in.
The time studies and the data housed in the IBF polygons are linked directly. A series of empirical constants result from the time studies. Through a 76-step equation, all of the relevant data housed in the IBF polygons are multiplied by the appropriate constants and reduced to equivalent beat design units (bdu) for each truck type and crew size. With this foundation data in place, the designer can now use the tools to design a beat.
Using Beat 2000, the city can answer questions such as, “How many trucks of this particular type and crew size will it take to service the work area if the city is hauling to District 3?” The system automatically determines the haul distances and appropriate bdu, and displays the answer for the designer. This system also allows the designer to analyze “on the fly” by varying the type of trucks and crew sizes, changing areas, selecting different district landfills as hauling destinations, and overriding tolerances. The result is a quick determination of the best mix of vehicles to use.
Once a vehicle or crew size is chosen, the designer then selects a serviceable area. Beat 2000 subsequently recalculates bdus for that area and displays the results. IBF polygons then are added or subtracted until the design matches the desired target values (the system continuously recalculates the design values based on polygon selection).
IBF polygon line work also is used to define the actual beat boundaries. Once a beat is defined, the system automatically generates an 8½-inch by 11-inch map for the truck driver's use in the field. If the size is too small to be easily read, the system automatically creates a readable size enlargement. As each beat is defined, the system automatically recalculates the remaining work area values.
Using this method, the summer 2000 design was created in 20 days compared to about one year using the city's previous methods.
All of the data and designs are interconnected through a complex series of networks and layers, which allow for multiple activities such as IBF polygon building, maintenance and different designs from different users all taking place simultaneously.
Assessing the Benefits
Once Beat 2000 was functional, the city's Solid Waste Services staff redesigned the entire city from scratch, including defining new district boundaries and changing all the existing beats. The first step was to maximize truck efficiency by analyzing the fleet and how the trucks were being used.
Next, each area of the city was examined, matching its needs to an appropriate truck type. The results for the summer period included using nine less vehicles on a regular basis compared to the previous year, despite a population increase of 18,361. For high-volume days, each district was able to add trucks from these newly created spares. In addition to the vehicle reduction, six fewer seasonal employees were required. Overall, substantial savings of approximately six figures were realized in fuel, maintenance and labor costs.
According to the city, the service day from the previous design remained the same, wherever possible. Each truck beat was assigned to a day and crew, and divided into foreman areas. These areas then were set to task force areas (an area that has to be completely serviced that particular day). Customers with a change in their collection day were informed using a hand-delivered notice.
To avoid any hauler conflicts, field input was solicited throughout the process, especially when determining the foreman area.
Each crew received the four sets of maps that corresponded to their four-day workweek. The same crews service the same areas for the entire summer.
Foremen reported the new beats' layout was easier to manage, and the maps were easy to follow. Drivers, too, liked the new maps and found the service locations to be useful. Crews adjusted to the new beat areas within a couple of weeks.
Overall, customers — except those with collection day changes — were not directly affected by the new beat designs. However, the city says customers will benefit from cost savings. Residential customers currently enjoy once-a-week refuse collection for less than $2 Canadian ($1.30 U.S.) per week. By using the new GIS system, this figure is expected to slightly decrease, despite the recent major increases in fuel costs and the city's continuing growth.
Cost savings resulted from being able to analyze the entire fleet to determine the best truck efficiencies based on actual scale house data, design criteria, capital and operating costs and overall performance.
Currently, Beat 2000 is being used to assess the impacts of changing from Monday through Thursday pickups to a Tuesday to Friday schedule. This change will benefit customers because their collection day would become permanent, instead of moving forward by a day after every long holiday weekend. Collection staff also are expected to benefit by eliminating the 10-day garbage load following holiday weekends.
Proposed changes and enhancements to Beat 2000 include rewriting the system in an Environmental Systems Research Institute Inc. (ESRI) environment; adding an Internet presence to allow customers direct access to collection information through the Web; adding scale-house data directly to the program; adding health and safety components and commercial container design capabilities; and including global positioning systems and onboard computers to the system.
So far, the city is pleased with the initial accomplishments of its Beat 2000 system. Calgary's Solid Waste Services Unit was recognized by the Solid Waste Association of North America, Silver Spring, Md., with a Gold Award of Excellence in collection for the year 2000.
With Beat 2000 providing a solid foundation upon which to build, the future holds even more promise.
Pete Truch is the project leader on Beat 2000, a system developed by and for the city of Calgary, Alberta, Canada, Solid Waste Services Business Unit. For more information on GIS and GPS systems, visit www.wasteage.com.