Comput., Environ.and Urban Systems,Vol. 20, No. 4/5, pp. 233-246, 1996 © 1997ElsevierScienceLtd All fights reserved.Printed in Great Britain 0198-9715/96$17.00 + 0.00 PH: S0198-9715(96)00019-1
A COMPUTER MODEL FOR SOLID WASTE INTEGRATED MANAGEMENT
F. S. Wang, * A. J. Richardson t and F. A. Roddicl~ •Department of Civil and Geological Engineering, Royal Melbourne Institute of Technology, Melbourne, Australia t Transport Research Centre, Royal Melboume Institute of Technology, Melbourne, Australia tDepartment of Chemical Engineering, Royal Melbourne Institute of Technology, Melbourne, Australia ABSTRACT. Solid waste management has evolved greatly since its early days and
it now considers an interrelated series of options aiming at waste source reduction, recycling, treatment and final disposal. A systems analysis approach is necessary when considering many options available and a systems model is desirable because of the interactions between many factors within a waste management system. Against this background, an interactive computer package called S W I M has been developed to provide a structure for systems analysis of solid waste management problems at the municipal level. It can assist decision makers to evaluate the economic and environmental impacts of various waste management options. The S W I M model was fully developed using Microsoft Excel 4.0 ® software on the Apple Macintosh platform by the end of 1994. It has now been further upgraded to Excel 5.0® for application on both IBM and Macintosh platforms. It incorporates a user-interactive and user-friendly interface which is necessary for such a decision support tool. This paper describes the modelling techniques, the interfaces and potential applications of the S W I M model. © 1997 Elsevier Science Ltd
INTRODUCTION Solid waste management, an indispensable part of our modern life on this planet, has become increasingly complicated since its earlier days. General awareness of our environmental problems has led to the development of pollution control technologies and more stringent legislation on waste handling and disposal to minimize the environmental impact associated with solid wastes. However, until recently the dominant practice of solid waste management has been an "end-of-pipe" approach. More recently, the desire to conserve our finite resources has shifted the central concern of waste management to waste prevention, minimization, and recycling (Hubick, 1991).
F. S. Wang et al.
Waste management now involves considering an interrelated series of options aiming at waste source reduction, recycling, treatment and finally disposal. However, environmental soundness is just one of the goals that decision makers have to consider in their solid waste management. To arrive at a better waste management system, they need to also consider their decisions in terms of economic viability and social equity. There is no doubt that economic resources are as critical to our sustainable development as our natural resources. Waste management needs to be economically viable to ensure that it is attainable in the short-term and sustainable in the long-term. Social equity is concerned with how equitably the benefits and costs of waste management are distributed among different groups within society. For example, although a regional waste treatment or disposal site may achieve the economics of scale required for economic viability, the siting of the facility may prove to be a most difficult issue. This is because the local community generally opposes having someone else's wastes dumped in their locality (the N I M B Y - - " N o t in my backyard" syndrome). In this case, the local community will be the loser while the regional community will be the winner. If this equity issue cannot be resolved (by compensating the loser), then wastes may have to be hauled further and the resultant waste management becomes economically less viable. Clearly, the best waste management strategies from either the environmental, or the economic, or the social equity point of view are not necessarily consistent. The cheapest way may not be the best environmentally. The best option in terms of social equity may not be the best economically or environmentally. These three aspects of waste management constrain, but also complement, each other in the development of a waste management system. It has become necessary for decision makers to take an integrated approach to consider not only a series of waste management options aimed at source reduction and environmentally sound practices, but also to integrate the environmental soundness with economical viability and social equity. To support this integrated approach to waste management, it is necessary to ascertain the environmental, economic and social impacts associated with various waste management options before decision makers can trade off these options to achieve a better waste management strategy. There are some hindrances, however, to such a determination. To name a few, there is the lack of information on waste generation and composition, inadequate pricing of waste management practices and a lack of systems analysis (Pearce & Turner, 1993). Among these hindrances, it is the lack of systems analysis that is most significant. The collection of information can only be justified by its usefulness in the decision-making process, and this will not be known until a systems study is carried out to investigate the critical information needed. Waste management practices, such as landfill disposal, cannot be correctly priced until they are assessed as parts of the entire waste management system in a dynamic way. For example, the pricing of the existing landfill disposal needs to take account of the future cost of not having a landfill site in the vicinity. This difficulty is even more of a problem when making a rational decision at the municipal level. This is simply because local councils are more limited in their economic and research resources for such a comprehensive systems study. However, the need for rational decision making at this level is the greatest as the overall impacts of the waste management system are closely related to their actual operations carded out at the local level. The local community is also the group most concerned with the economic, environmental and social impacts of a waste management strategy.
Computer models have been used in the solid waste management field for some time now. However, the previous developments were mainly concerned with the siting of waste treatment facilities (e.g. transfer stations) and the final disposal facilities (eg. landfill) and evaluations of a waste management strategy have been limited to only economic considerations. One example of such a model is LAWMAN (Local Authority Waste Management) by the CSIRO (Crawford, Mashford, Newton, & Sharpe, 1988). Therefore, there is a need to develop a decision-support aid to facilitate an integrated approach to waste management and to contribute towards rational decision making at the municipal level.
OBJECTIVE OF THE SWIM MODEL Against this background, the Solid Waste Integrated Management (SWIM) Model has been developed. The SWIM model was initiated in 1991 by the Faculty of Engineering at the Royal Melbourne Institute of Technology (RMIT) and the Transport Research Centre (then at the University of Melbourne). The SWIM model is an interactive computer package which provides a structure for a systems analysis at the municipal level to demonstrate the economic and environmental impacts of various options and so assist decision makers in the evaluation and design of waste management systems. The SWIM model has been applied to various waste management problems such as the impact of increased participation rates in a recycling program on the operations and costs of recyclables collection (Wang, Richardson, Roddick, & Curnow, 1994). It has addressed the economic and environmental implications of different ways of handling post-consumer paper, either by recycling using kerbside collection or central collection where paper is delivered by residents, or by dumping them in garbage stream (Richardson & Wang, 1994). It has also been applied in an investigation on the net cost incurred, or the profit generated, from weekly and fortnightly collections of recyclables (Wang, Richardson, & Roddick, 1996). It is the nature of simulation models to allow the exploration of complex systems in many different ways. The more variables (e.g. locations of facilities, size and type of collection trucks, type of recyclable materials to be collected) that users can specify, the more dimensions the model can investigate when simulating a complex system. The following list illustrates a few of the types of questions which can be addressed by the SWIM model: • What are the real costs associated with various collection services? • Is recycling economically viable in a particular local government area? • What collection system is more economically viable and/or more environmentally feasible (e.g. collection frequency, the types of recyclables and the sizes of trucks)? • What if a nearby landfill site is closed? Which option should be adopted: to use direct haul ot to use a waste transfer station? • How effective are recycling schemes for reducing wastes going to landfill? More importantly, how can an economically sustainable, environmentally viable and socially acceptable waste management system be achieved by considering the various options available?
F. S. Wang et al. METHODOLOGY
A systems analysis approach has been taken to develop the SWIM model. Constructing a valid conceptual model of the waste management system is the lrtrst step in exploring the system. Therefore, the interactions between various system components were identified by using a generic modelling package ithink @ 2.0 on the Apple Macintosh platform (Wang et al., 1994). Further development of the SWIM model was then performed using Microsoft Excel 4.0 @ software on the Apple Macintosh platform. Then, user-friendly interfaces were developed as an indispensable part of the model to ensure that SWIM users could easily explore the solid waste management system. Finally, the SWIM model has been verified and validated against operations in the former City of Nunawading, Melbourne. More specifically, in modelling the operational side of the solid waste management system, there are a few approaches adopted in the SWIM model which ensure that it can be tailored to each municipality. These approaches include: . the use of the Census Collector's District (CCD) as the basic unit of analysis; CCDs are used by the Australian Bureau of Statistics (ABS) for a variety of surveys and data reporting exercises. The use of the CCD as a basic unit enables substantial information, about the socio-demographics of the CCDs, to be utilized in the process of integrated waste management and planning. • the use of the x-y coordinates of CCD centroids as the spatial dimensions for analysis. The locations of waste facilities as well as the spatial dimensions of CCDs in any Local Government Area are specified using x-y coordinates. The capacity of the SWIM model is enhanced by this feature, particularly in terms of addressing problems relevant to specific geographical locations. Consequently, the SWIM model provides a decision-making aid for continuous improvement of solid waste management at the local municipal level.
THE SWIM MODEL
Major Components of the SWIM Model The SWIM model includes: • demand models that describe the demand for various waste management services in terms of waste generation rate, participation rate for recycling programs etc.; • supply models that relate the system's operating characteristics (e.g. collection costs and operational time) to the physical systems (e.g. locations of facilities and collection frequency) and to the demand for various waste and recycling services (e.g. set-out rates and participation rates); • impact models that evaluate economic impact in terms of the cost of providing the system, and the environmental impact in terms of the amount of C02 emissions from vehicles used in the collection and transportation processes; and
SWIM--A ComputerModel •
a user-friendly interface that makes data input to the model an easier process and enables the outputs from the model to be easily understood.
Major Features of the SWIM Model The major features of the SWIM model are that it: • is a simulation model; Simulation may be defined as the use of the computer to trace the lengthy chains of indirect repercussions in the cause-and-effect relations describing complex systems, and thus to imitate the dynamic behaviour of the system. The interactions between numerous variables, each of which can affect, and simultaneously be affected by, the others can hardly be amenable to purely mental evaluation or ordinary mathematical treatment. As a result, the use of a simulation model can greatly assist with understanding the complex solid waste management system that is a prerequisite for a sound decision-making process. • incorporates deterministic modelling with stochastic simulation techniques; By a deterministic modelling technique, the system is simulated based on average variables given exogenously. The outputs from the model are determined by the values assigned to the variables. The stochastic simulation, however, simulates a stochastic pattern in solid waste generation and collection processes and generates different outputs even when the same values are assigned to the variables. The deterministic model ascertains the different magnitudes of the interactions between various components, and the economic and environmental impacts associated with different designs of the waste management system. By contrast, the stochastic simulation, applied to daily operations of various collection services, offers insight into the collection systems with respect to capacity shortfalls and the economic implications of such shortfalls. The deterministic model is critical for an integrated and effective waste management system, and the stochastic simulation is important for efficient operation of each sub-system. The application of these two modelling techniques to a waste management system allows advancement of our understanding of the system and so more sensible decisions can be made thereafter. • is user-interactive and user-friendly; Great effort has been made for the SWIM model to be user-interactive and userfriendly. Dialog boxes, menu bars and a waste management system flowchart are available for SWIM users to specify numerous input parameters and to select various options (e.g. whether to have a transfer station or not). The task of defining a complex waste management system is thus greatly eased. An interactive and informative method for the design of collection areas is developed for allocating collection areas (a number of CCD units to be serviced by a truck on a collection day) to trucks carrying waste and recyclable materials. Users are informed when a collection truck is full (by weight or by volume), or when the operational time has filled the workday. Users can also select and de-select CCDs to find out the best combination for a collection area which is most consistent with operational constraints.
F. S. Wang et al.
A Few Examples of the Interface S W I M Menu T h e S W I M m e n u b a r consists o f five m e n u s , n a m e l y File, Design, Factors, Simulations a n d Outputs. Each m e n u c o n t a i n s a list o f m e n u items f r o m which users c a n elect to p e r f o r m v a r i o u s functions. T h e S W I M m e n u s a n d their m e n u items arc described briefly in T a b l e s 1 - 5. Each S W I M m e n u is s h o w n in bold. W h e n ellipses follow a m e n u item, it m e a n s t h a t w h e n this item is selected it will b r i n g o u t a further dialog b o x for m o r e instructions. T h e S W I M File m e n u has a similar l a y - o u t to the Excel File m e n u (Table 1). T h e m e n u item Exit to Excel offers users the o p p o r t u n i t y to use the s t a n d a r d Excel m e n u s . O n the s t a n d a r d Excel m e n u , a m e n u item S W I M is a d d e d to e n a b l e users to Table 1. File Menu Items and Their Functions Menu items
Create a new waste management (WM) project Open an existing WM project Close a WM project Save a WM project after it is specified Save outputs under different names Preview documents before printing, same as in the standard Excel menu Page set up, same as in the Excel menu Print outputs, same as in the Excel menu Exit to standard Excel working environment Quit from SWIM model and quit from Excel
New... Open... Close project... Save project Save AS... Print preview
Page setup... Print... Exit to Excel Quit
Table 2. Design M e n u Items and Their Functions M e n u items
Design Service types... Average house frontage... Bins per stop... Load factor... Contamination ratio... Frequency... Week... Check area Reset area...
Initialisc entire collection area
Bring out a dialog box for selectingone of the four kerbside service types for collection design Specify average household frontage Specify average number of bins per stop Specify the operational capacity which is a
percentage of the maximum capacity Specify the ratio of non-recyclables to recyclables in collections Specify collection frequency Allocate a collection area to a collection week, if collection is not a weekly collection Check if there are any CCDs missing in the collectionarea design Facilitate the trial and error procedure of designing collection areas by allowing reallocation of C C D s Clear the existing design and start a new design from a blank L G A m a p
choose the SWIM application from the standard Excel menu and to go back to the SWIM environment. This is advantageous since users can utilize many features of the Excel software to present outputs from the SWIM model in various ways. For example, graphical outputs can be created once tabulated outputs from the SWIM model have been generated. The Design menu includes ten menu items (Table 2). It is used for the design of collection areas based on the Local Government Area (LGA) maps, which will be described in detail later. The initial state of the Design menu will only have the item "Service Types..." available. Once a type of kerbside collection service (e.g. garbage collection) is chosen, the SWIM model changes to the design setting which shows SWIM LGA map(s) and the accompanying custom toolbar for designing collection areas. The other menus then become available to assist in the allocation of CCDs to collection trucks and to collection days. For example, the menu item Load Factor allows the user to specify the percentage of the maximum truck capacity (by volume and by weight) utilized under normal conditions. If the load factor is 80%, the model will alert the user when the material collected from CCDs has filled 80% of the truck capacity. A collection system designed in this manner will be able to accommodate demand fluctuations based on seasonal and random fluctuations. Custom toolbars with design-related features are also developed and will be described later. The Factors menu includes five menu items which are described in Table 3. These menu items are for entering unit costs for various waste management operations. In the SWIM model, costs for collecting waste or recyclables from households, costs for transporting waste to disposal sites and costs for hauling recyclables to recyclable reprocessing industry sites are simulated by the supply system models. However, costs for operating transfer stations, for composting and mulching at central sites, for sorting recyclables and for landfill disposal, and market prices for recyclables must be specified by the users. The Simulations menu includes four menu items for simulating four types of collection services (Table 4). Selecting any of the four menu items will bring forth a dialog box presenting two simulation options: average performance evaluation or daily simulation. The option of average performance evaluation represents the deterministic modelling and the option of daily simulation represents the stochastic simulation. An example of this dialog box will be given later. The Outputs menu consists of seven menu items for various outputs from the SWIM model (Table 5). It can be seen that the SWIM menus provide a wide range of functions Table 3. Factors Menu Items and Their Functions Menu items
Factors Sorting costs... Prices... Waste transfer cost... Composting/mulching cost... Landfill charges...
Specify unit sorting cost for sorting comingled recyclables Specify market prices for various recyclable items Specify unit operating cost of a waste transfer station Specify unit operating cost of a composting/ mulching operation Specify landfill charge
F. S. Wang et al.
Table 4. Simulations Menu Items and Their Functions Menu items
Simulations Garbage collection operations... Recycling collection operations... Paper cotlection operations... Green organics collection operations...
Select simulation options for a garbage collection service Select simulation options for a co-mingled recyelables collection service Select simulation options for a paper collection service Select simulation options for a green organics coUection service
that assist the user to specify some of the variables, to simulate the waste management options and to generate outputs from the SWIM model. Apart from the SWIM menus, a graphic interface, i.e. the SWIM system flowchart, is used for specifying many other variables and another graphic interface, i.e. the SWIM LGA maps, is used for interactive design of collection areas.
SWIM System Flowchart The SWIM system flowchart is a graphic interface. Users define a system by choosing the File~New menu item and open a waste management system flowchart as shown in Figure 1. A system flowchart is used for entering four types of information about a waste management system: Table 5. Outputs Menu Items and Their Functions Menu items
Outputs Waste streaming summary Recyclable quantities and revenues Recycling rates Performance of garbage collection Performance of co-mingled recyclables collection Performance of paper recycling Performance of green organics recycling Collection maps... Waste transfer and haulage Recyclables haulage Waste management summary
Summarize the quantifies of materials in garbage collection, various recycling schemes and source reduction activities Snmmarize the annual quantities of recyclables collected and the revenues Report on the rate of a recyclable collected to the total weight of the recyclable available (by weight) Brings out the tabular output from the deterministic simulation for garbage collection Brings out the tabular output from the deterministic simulation for co-mingled reeyclables collection Brings out the tabular output from the deterministic simulation for paper recycling Brings out the tabular output from the deterministic simulation for green organics recycling Brings out a dialog box with options for choosing maps for one of the four kerbside services Calculates cost and CO2 associated with waste transfer and bulk haul operation Calculates cost and CO2 associated with hauling reeyclables to recycling industry sites Summarizes costs for various waste management services and private transportation of wastes
Design Factors Simulations
flowch~t FIGURE I. An example of a SWIM system flowchart.
• the information on waste generation, waste composition and flow of wastes generated from the domestic source into various collection services and activities; • the specifications of collection services in terms of the bin used, the number of trucks, the capacity of the truck and the operational information for the trucks; • a measure of households using the services, in terms of the participation rate, and the frequency of usage in terms of set-out rate; and • the geographic locations of various waste-handling facilities such as transfer stations. Each picture in a system flowchart is a function button which has bccn assigned to Excel programs for various data entry tasks. Clicking on the pictures will bring forth dialog boxes whereby the user can enter the data. The data are then stored in the files created by the model for later uses. The graphic lay-out of the solid waste management system greatly eases the daunting task of specifying the system in terms of the waste generation, storage, collection, collection vehicles, waste-handling facilities, etc. For example, clicking on the "house" picture produces the dialog box shown in Figure 2(a). Two options are available for the user to specify waste generation rates. Selecting "Average generation rate" brings forth another dialog box which allows the user to specify an average waste generation rate for all types of dwellings (Figure 2(b)).
F. S. Wang et al.
FIGURE 2. (a) Dialog box for spec/fyiog waste generation rate. Co) Dialog box for specifyi~ average waste generation rate. (e) Dialog box for specifying variable waste generation rates.
Selecting "Variable generation rates" brings forth a different dialog box (Figure 2(c)) which allows the user to specify variable waste generation rates by dwelling types. In most of the dialog boxes, default values from the literature are supplied, such as the average waste generation rate shown in Figure 2(b). However, where literature is not available, estimates are used as initial data, such as the variable waste generation rates shown in Figure 2(c). As shown in the above example, the entry of data is made easy and straightforward by dialog boxes. It is also flexible in that it allows both an aggregated or disaggregated approach to specify some of the important system variables.
Collection Area Design A SWIM LGA map is a map of the LGA which has been divided into CCDs and is shown on-screen in the process of designing collection areas. There may be more than one such map for an LGA, depending on the geographic features of the LGA. For example, Figure 3 shows a SWIM LGA map for a part of the City of Clarence, Tasmania, Australia,
Design Factors Simulations Outputs
FIGURE 3. Exampleof a SWIM LGA map for the city of Clarence. where the SWIM model has been applied to assist in the development of an integrated waste management system. On this map, an oval button inside a CCD boundary represents the CCD. Each CCD is characterized by its centroid in x - y coordinates, the number of dwellings, the amount of garbage and recyclables to be collected, the participation rates for various recycling services (a participation rate for a recycling program is defined as the percentage of households which put out recyclables for collection at least once over a period of time, e.g. six weeks) and the set-out rates for collection services (a set-out rate is defined as the average percentage of households that puts out materials on a collection day). The custom design toolbar appears only when one of the L G A maps (if more than one) is activated. Some custom design toolbars are used for all types of collection services while the others are specific to each type only. The toolbars for garbage collection design are shown in Figure 3 on top of the LGA map. The first toolbar on the far left side is the same for all services, and it is used for manipulating the maps. It has tools for zooming in and out (the first and second tools). Tools for activating L G A maps can be included, as in this example for Clarence: the third tool of the first toolbar is for activating the whole Clarence map; the fourth is for activating the Clarence inner map. The second toolbar is also the same for all services, and is used for specifying collection day and selecting a truck. However, the third toolbar will appear differently for different services, indicating various waste/reeyelables handling facilities.
F. S. Wang et al.
When a CCD button is clicked, the information about the CCD will be displayed and used by the SWIM model to calculate the information related to the progress of the collection design. The information is displayed in a small window, called "Areainfo", which is located next to the LGA map on-screen to assist the user in the design of the collection areas (Figure 4). In this window, information on four parameters is given. These are: • • • •
the the the the
number of dwellings serviced weight of materials collected (tonnes) volume of materials collected (cubic metres) time taken to collect materials from households (hours).
The information is given at three different levels, which are "This CCD", "Run total", and "Daily totals". A typical sequence of runs in a workday is shown schematically in Figure 5. Run 1 begins from the time that a vehicle leaves the truck depot to collect waste from CCDs to the time the vehicle unloads at an unloading site. If the collection area for that truck, for a workday, has not been serviced fully, run 2 starts when the vehicle leaves the unloading site to service another group of CCDs and lasts until the time that the vehicle unloads at the unloading site. If there is still part of the collection area yet to be serviced, run 3 starts when the vehicle leaves the unloading site, services the remaining CCDs, unloads at the unloading site and returns to the truck depot at the end of the workday. In the "Areainfo" window, the time given for the current CCD is solely the collection time taken to collect materials from households and move between dwellings in this CCD. The time given for the "Run total" is the total time spent so far in all the CCDs serviced in a run. The time given for the "Daily totals" includes the collection time in all runs and the access time needed, from the truck depot to the collection areas, and from the collection areas to the unloading sites. The information displayed in the "Areainfo" window corresponds to the progress of the collection design. For example, in the City of Nunawading, garbage collection trucks unload at the Nunawading Transfer Station. When the user clicks on the "Transfer Station" button to indicate unloading of a collection truck, the information about the
FIGURE 4. The "Aresinfo.xlw" window.
S W I M u A Computer Model
FIGURE 5. A typical sequence o f runs witl~ a workday.
previously selected CCD in the column headed "This CCD" is cleared, but "Run total" and "Daily total" information remains as a summary of the previous run and the daily total. By observing the information in the "Areainfo" window, the user can decide when to finish a run, either by the capacity used (e.g. the volume used in the truck) or by the time elapsed from the start of the day. Thus the design of collection areas is user-interactive as well as informative.
Simulation Options The SWIM model not only offers great ease and flexibility for the user to specify exogenous variables such as the waste generation rate, it also offers deterministic modelling as well as stochastic simulation for the user to explore the waste management system. For example, selecting one of the Simulations menu items produces the dialog box shown in Figure 6.
FIGURE 6. Options for sulmnlmfinga waste eoHeelion system.
F. S. Wang et al.
By choosing the option of "Average performance evaluation", the SWIM model simulates a collection system using the deterministic modelling technique where the outputs are determined by the various inputs. Consequently, the magnitude of the problems concerned (e.g. collection costs) can be ascertained. By contrast, choosing the option of "Daily simulation" causes the SWIM model to apply different seasonal factors to the quantity of materials to be collected and random factors to the number of bins set out on collection days. Hence the stochastic patterns of waste generation, and of collection operations are simulated. The simulation outputs are then tabulated, and graphs can be created from the tabulated outputs using the standard Excel charting tool. As a result, collection services are investigated in depth in order to determine any capacity shortfalls in daily operations. When capacity shortfalls are uncovered by the daily simulations, decisions can be made to improve the design of the collection system so that better operational management can be rendered.
CONCLUSIONS This paper has described the development of the Solid Waste Integrated Management (SWIM) m o d e l - - a simulation computer m o d e l - - for decision support at a local level. The system analysis approach, the features and the interface of the SWIM model have been presented. Moreover, discussion has demonstrated that the SWIM model is capable of simulating a complex waste management system in an interactive and user-friendly way, thereby assisting decision makers to evaluate various waste management options in terms of their economic, environmental and social impacts. In short, solid waste management can be enhanced in its effectiveness, efficiency, overall economical viability and environmental feasibility through the use of the SWIM model. ACKNOWLEDGEMENTSmThe authors thank the Resource Recovery and Recycling Council for partial funding to assist this project. The authors also acknowledge the assistance provided by council officers from the former City of Nunawading, Melbourne, especially Andrew Mcintosh, Graham Hawke, Bret Jones and John Prince.
REFERENCES Crawford,J. R., Mashford,J. S., Newton,P. W., & SharpeR. (1988).A planningmodelfor solidwastedisposal. In P. W. Newton, M. A. P. Taylorand R. Sharpe (Eds.), Desktop planning: microcomputer applicationsfor infrastructure and services planning and management (pp. 155-165). Melbourne:Hargreen. Hubick, K. T. (1991). Management and technologies of wastes: A perspective--Australia 1990. Departmentof Industry,Technologyand Commerce,February. Pearce, D. W., & Turner, R. K. (1993). Market-based approaches to solid waste management. Resources, Conservation and Recycling, 8, 63-90.
Richardson, A. J., & Wang, F. S. (1994). DomesticSolidWasteManagement:the ForgottenEnd of the Urban Goods MovementChain. ForumPapers, 19th Australasian Transport Research Forum, (pp. 583-601). Wang, F. S., Richardson,A. J., & Roddick,F. A. (1996). Developmentand Applicationof the SWIMModel.3rd National Hazardous and Solid Waste Convention, (pp. 111-118). Sydney,May. Wang, F. S., Richardson,A. J., Roddick,F. A., & Curnow, R. C. (1994). SWIM--Interactivesoftware for continuousimprovementof solidwastemanagement.Journal of Resource Management and Technology, 22(2), 63-72.