A GIS based transportation model for solid waste disposal A case study on Asansol municipality

M.K. Ghose
b

a,*

, A.K. Dikshit b, S.K. Sharma

a Regional Remote Sensing Service Center, ISRO, IIT Campus, Kharagpur 721 302, India Centre of Environmental Science and Engineering, Indian Institute of Technology, Bombay, India c Department of Civil Engineering, Indian Institute of Technology, Kharagpur, India

Abstract Uncontrolled growth of the urban population in developing countries in recent years has made solid waste management an important issue. Very often, a substantial amount of total expenditures is spent on the collection of solid waste by city authorities. Optimization of the routing system for collection and transport of solid waste thus constitutes an important component of an eective solid waste management system. This paper describes an attempt to design and develop an appropriate storage, collection and disposal plan for the Asansol Municipality Corporation (AMC) of West Bengal State (India). A GIS optimal routing model is proposed to determine the minimum cost/distance ecient collection paths for transporting the solid wastes to the landll. The model uses information on population density, waste generation capacity, road network and the types of road, storage bins and collection vehicles, etc. The proposed model can be used as a decision support tool by municipal authorities for ecient management of the daily operations for transporting solid wastes, load balancing within vehicles, managing fuel consumption and generating work schedules for the workers and vehicles. The total cost of the proposed collection systems is estimated to be around 80 million rupees for the xed cost of storage bins, collection vehicles and a sanitary landll and around 8.4 million rupees for the annual operating cost of crews, vehicles and landll maintenance. A substantial amount (25 million rupees/yr) is currently being spent by AMC on waste collection alone without any proper storage/collection system and sanitary landll. Over a projected period of 15 yr, the overall savings is thus very signicant.

1. Introduction Technological development, globalization and population growth have accelerated the dynamics of the urbanization process in developing countries. In India alone, the urban population has increased from 11% in 1901 to 26% in 2001. The rapid growth rates of many cities, combined with their huge population base, has left many Indian cities decient in infrastructure services like water supply, sewerage and solid waste management.

Due to a lack of serious eorts by town/city authorities, the management of garbage has become a tenacious problem, notwithstanding the fact that the largest part of any municipal expenditure is allotted to it. It is estimated that the urban local bodies spend about 500 rupees (1 US$ = 45 rupees approx.) per ton on solid waste collection, transport and disposal, which may rise to 1500 rupees per ton in some instances. A substantial amount of the total expenditure (85%) is spent on collection and as such improvement in the design of the collection systems could result in substantial savings, thereby saving a large proportion of the funds. However just collecting the waste from dierent parts of city does not solve the problem, it requires disposing the waste in environmentally safe and economically sustainable

1288

manner. An eective solid waste management system is needed to ensure better human health and safety. In general, an eective solid waste management system should include one or more of the following options: waste collection and transportation; resource recovery through waste processing; waste transportation without recovery of resources, i.e., reduction of volume, toxicity, or other physical/chemical properties of waste to make it suitable for nal disposal; and disposal on land, i.e., environmentally safe and sustainable disposal in landlls (Tchobanoglous et al., 1993; Kreith, 1994; CPHEEO, 2000). Most often, it appears dicult to minimize two variables cost and environmental impact simultaneously. Hence, the balance that needs to be struck is to reduce the overall environmental impact of the waste management system as far as possible within an acceptable cost limit. In this paper, an attempt has been made to propose an ecient solid waste management system for municipal solid waste, excluding industrial, constructional and hospital waste. The framework of the proposed waste management plan is mainly based on the following considerations:     Appropriate method of on-site storage. Appropriate method of bulk storage of waste. Appropriate method of primary collection of waste. Appropriate method of transportation of waste using Geographical Information System (GIS).  Appropriate method of waste disposal.  Financial expenditure on whole solid waste management plan. Development of an optimal routing scheme for waste disposal involves determination of a number of selection criteria, which is a tedious job for a planner to do manually. Various optimization models are discussed in the literature. Pelms and Clark (1971) proposed a location model for optimal location allocation, while Male and Liebman (1978) used the districting and routing for solid waste collection. Clark and Gillean (1974) gave a system analytical approach, and Baetz (1990) adapted simulation modeling for optimal solid waste collection. In the present study, the principles of the Geographical Information System (Heywood et al., 1988) are mainly emphasized in planning an eective solid waste management system. 2. Description of the study area and the existing waste management system The proposed solid waste management system is designed for the Asansol Municipality Corporation (AMC). Asansol (Longitude 87E and Latitude 2340 0 N) is an industrial town covering an area of 127.24 km2 situated in the Barddhaman District of the state of West Bengal, India. The Asansol Municipality has 50 wards and 95,293 households, with an average number of occupants per house of 5. It is estimated that the city produces 180 MT of total solid waste daily, which includes 60 MT of

Table 1 Solid waste data of Asansol City Particulars of solid waste Solid waste generation rate Total quantity of solid waste generated Quantity of domestic solid waste Total number of community bins Total annual expenditure for SWM Total number of supervisory sta Total number of sub-ordinate sta Value 0.250 kg/capita/day 180 MT/day 120 MT/day 1350 Rs 25 million 20 487

Source: National Institute of Urban Aairs, Ministry of Urban Development, Govt. of India.

non-domestic waste such as industrial, construction and hospital waste for which the owners of the respective organizations have responsibility for managing. Asansol Municipal Corporation (AMC) is solely responsible for collection and disposal of 120 MT/day of domestic solid waste. Presently, the existing 1350 masonry dustbins at selected places are used for primary storage. The waste materials are transported using 200 hand carts, 23 trucks and 2 tractors for disposal in open space at the outskirts of the city. About 25 million rupees annually are expended for the entire solid waste disposal process, without any proper collection system or sanitary landll. The detailed supportive data on waste collection statistics of Asansol city is given in Table 1. 3. Proposed GIS based solid waste collection system The proposed GIS model for solid waste disposal involves the planning of bins, vehicles and optimal routing. The generation of a spatial database on the road network, bin locations, landll site and garage uses collateral information and updating the same using satellite data. 3.1. Storage of waste Based on population density, width of roads, availability of space and minimum travel distance from a house, three types of bins viz. type-A (7.0 m3), type-B (0.75 m3), type-C (0.5 m3) have been proposed to be placed in dierent parts of the city. Some extra capacity is being provided for unforeseen factors such as slippage, overow, etc., and the container utilization factor is assumed to be 50%. All roads in AMC have been classied into three categories: major roads (width: 57 m), minor roads (width: 2.5 5 m) and other roads (width: less than 2.5 m). The road network, along with the ward boundaries and the administrative boundary, of AMC is shown in Fig. 1. It has been estimated that about 55 bins of type-A, 570 bins of type-B and 780 bins of type-C will need to be placed at dierent parts of the city along its major, minor and other roads. It is assumed that the frequency of collection for bin type-A is daily, but for bin type-B and bin type-C collection is every second day of the week as shown in Table 2. The spatial distribution of storage bins over the road network of AMC is shown in Fig. 2.

1289

Fig. 1. Road network and ward/administrative boundaries of AMC.

Fig. 2. Spatial distribution of storage bins over the road network of AMC.

3.2. Collection of waste Collection of solid waste is carried out by using suitable vehicles. The type of vehicle to be used depends on the type of collection bin and width of road (Chiplunkar et al., 1981). Hence, three types of vehicles are used for the three types of bins. All of the mechanised vehicles are chosen to reduce the pick-up time of bins at dierent locations and thus to reduce the number of vehicle requirements. The three types of vehicles designed for managing the solid waste in AMC are as follows: Vehicle Type-A: It is a skipper type of vehicle having a length of 4.5 m. It lifts only A-type bins and travels only on the major roads. It can carry only one bin at a time. Vehicle Type-B: It is a lifter type of vehicle with a frontloading mechanism and lifts B-type bins and travels both on major and minor roads.
Table 2 Conguration of dierent types of bins Bin type A B C Volume (m3) 7.00 0.75 0.50 Population served >300 300 200 Sources of waste generation Market places, street vendors and C-type bins Multi-storied buildings, commercial complexes, community and other private societies Slums and congested areas Total No. of bins 55 570 780 Period of lling 1 day 2 days 2 days Frequency of clearance Every day Every second day Every second day

Vehicle Type-C: It is an auto-rickshaw type of vehicle and is used for the collection of waste from the congested areas. It can collect wastes from C-type bins and unload it into the nearest A-type bin. The collection congurations of various vehicles planned for the present study is given in Table 3. The congurations of different types of vehicles and those of the storage bins are shown in Figs. 3 and 4.

3.3. Collection methodology adopted Only one landll and one garage are assumed for the present study. The sanitary landll has been designed for an active life of 15 yr and post closure height of 20 m. Considering the provision for infrastructures viz. site fencing all around

Container utilization factor = 50%. Extra capacity is being provided for unforeseen factors.

1290 Table 3 Collection conguration of vehicles Vehicle type A B C Volume (m3) 20 (4 2.5 2) 7.5 (2 1.5 2.5) Crew 2 2 2 Bins collected in a trip 1 50 28 Estimated waste collected in a trip (at actual utilization of bins) 50 0.75 0.5 = 19 28 0.5 0.5 = 7

Fig. 3. Congurations of dierent types of storage bins.

Fig. 4. Congurations of dierent types of collection vehicles.

the landll, two weigh bridges of 50 ton capacity, an administration oce of 30 m 10 m, a site control oce of 3 m 2 m (potable oce), access roads (all main roads

are 7 m wide and all arterial roads are 3.5 m wide) all around the periphery, the total area needed for the landlling operation has been estimated as 414.5 103 m2. For

1291

managing solid waste each day, a daily cell of size 30 m 22 m with a lift of 2 m will be used. The naturally available soil cover of 20 cm thickness will be provided on each cell daily. The volume of soil needed as the cover for the cell will be 132 m3/day. After the end of each phase, i.e., 1 yr, all surfaces will be covered with 40 cm thick naturally available soil. All of the essential components of a landll, including the liner system, leachate collection system, gas collection system, nal cover system, surface water drainage system, environmental monitoring system and closure and post closure plan have been designed and are discussed elsewhere (Sharma, 2002). Normally, daily working hours for a crew is 8 h, including time of lunch and clearance at the landll. Each vehicle consists of two crew members who are fully responsible for the collection and disposal of the wastes from the bins. The speed limit is assumed to be the same for all vehicles of the same type. However, the GIS router model has the option to input the user dened speed limit for respective types of vehicles. Moreover, delays due to trac jams, one-way roads, etc., can be input through link/arc-impedances of the proposed GIS-model. The vehicle type-A starts from the garage; travels to the location of the type-A bin; transports the bin to the landll and then returns back to previous place to replace the bin. Then the vehicle moves to the next nearest type-A bin and repeats the same process. At the end of the day, the vehicle will return to the garage after replacing the last bin served, at its original location. The vehicle type-B starts from the garage, collects the waste from 50 B-type containers and moves to the landll. The vehicle will then come to the next B type bin location and repeat the process. At the end of the day, the vehicle will return to the garage from the landll. The vehicle type-C starts from the garage, collects the waste from 28 C-type bins and disposes the waste at its nearest A-type bin. The vehicle will then come to the next C-type bin location and repeat the process. At the end of the day, the vehicle will return to the garage from the location of the last A-type bin served. 3.4. Optimal GIS routing model for collection and disposal of solid waste Collection routes are worked out by using GIS (Heywood et al., 1988). The NETWORK module (ESRI, 1995) of Arc/Info GIS software is used with the planned infrastructure to nd the shortest or minimum impedance path through a network. The speed of the vehicle is taken as the arc impedance and no turn impedance is being used. For the present study, the speed limit for all the vehicles are assumed to be the same; however, the user dened speed limit for respective types of vehicles can be fed into the model as the values of arc impedances. Moreover, delays due to trac-jams and signals, one-way roads, etc,. can be input through link/arc-impedances of the proposed model. The time spent at the landll and at the bins for

clearance of the wastes is taken as the node/stop-impedance and it can vary from bin to bin. The time taken to travel each line segment is calculated by the speed of the vehicle and the road type. Three types of ARC/INFO NETWORK modules named as GIS-router modules have been developed using AML (Arc Macro Language) depending upon the type of collection methodology to facilitate the modeling of spatial networks. The shortest path computed by using a heuristic procedure viz. the traveling salesman problem (Sharma, 1974) is saved as a route in a route-system and displayed using the route display commands in ARCPLOT. The NETWORK commands PATH and TOUR are used for nding minimum paths. 3.5. Optimal routing of transport system A-type vehicle: The major roads and the A-type bins are selected from the spatial data. The order in which the bins have to be visited is calculated based upon the proximity of bins. This initial planning can also be edited later. Then the optimal path is generated for each vehicle. In this process, clusters of bins are formed and each cluster is allocated to a vehicle. The clusters are made by taking the time into account, which may be plus or minus the total working hours for the day. The last cluster may need less time if the number of bins to be cleared is less. In such cases, the last cluster may be merged with other clusters by increasing the total working hours. B-type vehicle: The ordering of the B-type bins is performed as done for the A-type vehicle. Based upon the ordering, clusters of 50 bins are formed, which is the maximum capacity that a B-type vehicle can clear at a time. The total time required to clear each cluster through the optimal path is calculated and displayed to the user. Now the user can select a subset of clusters to be cleared by a vehicle. Once the user enters the group of clusters to be cleared, a nal optimal path between these clusters is generated. The total time required and other results are calculated. C-type vehicle: Here all of the A-type bins that are having links to C-type bins are selected. The visiting order of A-type bins is calculated. Then the optimal path is calculated for each C-type cluster attached with A-type bins. Based upon the total working hours, the number of vehicles required to clear the C-type bins is then calculated. 4. Results and discussion 4.1. Clearance of A-type bins The user is provided with the various optimum paths that an A-type vehicle has to take for clearing A-type bins in a given time. Each vehicle is provided with the cluster of bins and the order in which it has to be cleared. The number of trips required, time and total number of km each vehicle has to run is also calculated. Based on the number of vehicles available, the user can generate a schedule for

1292

vehicles and wages for workers. For example, a sample output of the optimal path for one A-type vehicle is shown in Fig. 5. The travel time, distance covered, and number of bins cleared by A-type vehicle are shown in Table 4. 4.2. Clearance of B-type bins As one B-type vehicle can clear a group of B-type bins at a stretch, the user is initially given the estimated time for clearing the clusters of B-type bins. From the set of clusters, the user can schedule a vehicle for a day. Once the input is given by the user, all of the results as obtained for B-type bin are generated. The travel time, distance covered and number of bins cleared by the B-type vehicle is shown in Table 5.

Table 5 Travel time, distance covered and number of bins cleared by type-B vehicle Day of clearance Day 1 Vehicle No. 1 2 3 1 2 3 1 2 3 3 Travel time (h) 9 (8.76) 9 (9.17) 7 (7.31) 8 (8.41) 7 (7.37) 8 (7.90) 8 (7.61) 9 (8.77) 9 (8.75) 74 Distance traveled (km) 98 (98.13) 94 (93.48) 74 (73.65) 91 (91.40) 77 (77.20) 95 (95.27) 76 (75.80) 97 (97.34) 98 (97.45) 800 No. of B-type bins cleared 200 200 170 200 170 200 170 200 200

Day 2

Day 3

Total

4.3. Clearance of C-type bins Here the waste from C-type bins is cleared by a C-type vehicle and disposed in the designated A-type bin. The same set of results as in the case of A-type bin is generated. The travel time, distance covered and number of bins cleared by a C-type vehicle is shown in Table 6. The total number of dierent types of vehicles, distance traveled and total working hours as calculated in the GIS
Table 6 Travel time, distance covered and number of bins cleared by type-C vehicles Day of clearance Day 1 Vehicle No. 1 2 3 4 5 6 1 2 3 4 5 6 1 2 3 4 5 6 6 Travel time (h) 7 8 7 7 6 6 7 7 6 6 8 8 6 6 8 7 7 8 (7.21) (7.91) (7.40) (6.89) (6.31) (5.87) (6.63) (7.11) (6.27) (6.13) (8.06) (7.88) (6.39) (5.95) (7.84) (7.36) (7.25) (8.12) Distance traveled (km) 41 59 54 39 36 32 36 54 37 35 44 39 33 29 52 47 47 61 (41.13) (58.45) (54.37) (39.11) (35.98) (31.63) (36.29) (53.46) (36.78) (35.27) (44.22) (39.44) (33.43) (29.08) (51.68) (47.17) (47.00) (60.72) No. of C-type bins cleared 140 140 140 140 112 108 140 140 112 108 140 140 112 108 140 140 140 140

Day 2

Day 3 Fig. 5. A sample output of the optimal path for one A-type vehicle. Table 4 Travel time, distance covered and number of bins cleared by type-A vehicles Vehicle No. Time of travel (h) 7 7 7 7 7 6 (7.20) (6.91) (6.58) (7.16) (7.09) (6.04) Distance traveled (km) 116 (116.26) 95 (94.45) 97 (97.40) 78 (77.45) 76 (75.45) 81 (81.25) 543 Number of bin type-A cleared in a day 8 9 8 11 11 8

Total

125

775

Table 7 Details of collection and transportation system Types of vehicles A B C Total Number of vehicles required 6 3 6 15 Total distance traveled (km/week) 3801 800 775 5376 Total working hours (h/week) 370 74 125 569

1 2 3 4 5 6 Total = 6

41

1293 Table 8 Estimated xed cost for the solid waste management plan of AMC S.No. 1 2 3 4 5 6 7 Total Item Bin type-A Bin type-B Bin type-C Vehicle type-A Vehicle type-B Vehicle type-C Landll site construction Quantity 55 570 780 6 3 6 1 Unit cost in rupees 15,000 9000 2500 4,500,000 6,500,000 100,000 Total cost in rupees 825,000 5,130,000 1,950,000 27,000,000 19,500,000 600,000 25,000,000 80,005,000 Approx. 80 million

the AMC is spending around 25 million rupees annually in operating costs for collection alone without any proper storage/collection system and sanitary landll. 6. Conclusion In this paper an attempt has been made to design and develop an appropriate storage, collection and disposal plan for the Asansol Municipality Corporation (AMC) of West Bengal State (India). A GIS optimal routing model based on the parameters such as population density, waste generation capacity, road network and the types of road, storage bins and collection vehicles, etc., is developed and used to trace the minimum cost/distance ecient collection paths for transporting the solid wastes to the landll. The proposed model can be used as a decision support tool by the municipal authorities for ecient management of the daily operations for moving solid wastes, load balancing within vehicles, managing fuel consumption and generating work schedules for the workers and vehicles. References
Baetz, Brian W., 1990. Optimization/simulation modeling for waste management capacity planning. Journal of Urban Planning and Development 88, 5979. Chiplunkar, A.V., Mehndiratta, S.L., Khanna, P., 1981. Optimization of refuse collection systems. Journal of Environmental Engineering Division, ASCE 107 (EE6), 12031211. Clark, Robert M., Gillean, James I., 1974. System analysis and solid waste planning. Journal of Environmental Engineering Division, ASCE, 7 23. CPHEEO (Central Public Health and Environmental Engineering Organization), 2000. Manual on Municipal Solid Waste Management. ESRI, 1995. Network Analysis-Modeling Network Systems. Environmental Systems Research Institute Inc., Redlands, CA, USA. Heywood, I., Cornelius, S., Carver, S., 1988. An Introduction to Geographical Information Systems. Addison Wesley Longman, New York. Kreith, F., 1994. Handbook of Solid Waste Management. McGraw Hill, New York, USA. Male, J.W., Liebman, J.C., 1978. Districting and routing for solid waste collection. Journal of Environmental Engineering Division, ASCE 1, 114. Pelms, Billy P., Clark, Robert M., 1971. Location model for solid waste management. Journal of Urban Planning and Development 102, 129. Sharma, S.D., 1974. Operations Research. Kedar Nath Ram Nath & Co., Meerut. Sharma, S., 2002. Developing an integrated solid waste management plan for Asansol city, M.Tech. Thesis, Department of Civil Engineering, IIT Kharagpur. Tchobanoglous, G., Theisen, H., Vigil, S.A., 1993. Integrated Solid Waste Management: Engineering Principles and Management Issues. McGraw Hill, Singapore.

Table 9 Estimated annual operating cost for the proposed solid waste management plan of AMC S.No. 1 Item Vehicles running cost including the maintenance Labor cost in collection Landll running cost Unit cost in rupees 20/km Total cost in rupees 5,591,040 (5376 52 20) 1,800,000 (15 2 5000 12) 1,000,000 8,391,040 Approx. 8.4 million

2 3 Total

5000/crew

environment for the whole solid waste management system is given in Table 7. 5. Financial expenditure Being a public utility and an essential service, investment in solid waste management does not require a justication in terms of positive return on investment or minimum prots. Such an investment, however, needs to be justied on the basis of being the least cost technologically feasible option for achieving the required degree of eciency. Disposal cost can be dened as total cost incurred by AMC in disposing of the municipal solid waste and, accordingly, the estimated xed and operating cost for solid waste management are obtained as shown in Tables 8 and 9, respectively. The total cost for the newly designed collection systems is estimated to be around 80 million rupees for the xed cost of storage bins, collection vehicles and a sanitary landll and around 8.4 million rupees for the annual operating cost of crews, vehicles and landll maintenance. Presently,