Route optimization for solid waste collection: Trabzon (Turkey) case study

Abstract

In a municipal solid waste management system, decreasing collection/hauling costs, which consist of 85 % of total disposal expenditure, can be carried out by a route optimization. Thus, a huge amount of economical benefits is getting furnished. If route optimization is performed in solid waste collection/hauling process, due to reductions in "empty miles" negativity, total expenditures will be decreased. Trabzon City located in the northeast side of Turkey has about 185 thousand inhabitants according to Census 2000. The city shares just about 1% of the Gross Domestic Income in Turkey. In other words, that means that Trabzon City livings have moderate revenue. The objectives of this study are to optimize for the route of collection/hauling in Trabzon City by taking consideration of data about road net, demographics and solid waste production. In order to analyse the solid waste collection/hauling process in the city, the processes were recorded by a Sony DCR-TRV145E brand video camera. To use route optimization process, data related in present spending, truck type and capacity, solid waste production, number of inhabitants and Global Positioning System (GPS) receiver data for each route were collected and all the data were analyzed with each other. For 39 districts in the city, a shortest path model was used in order to optimize solid waste collection/hauling processes, as minimum cost was aimed. The Route View Pro TM software as an optimization tool was used for that purpose. Geographic Information System (GIS) elements such as numerical pathways, demographic distribution, container distribution and solid waste production amount were integrated to the software. To give an idea, thematic container layer has 777 container location points for the entire city. After performing routes by the software, the optimized routes were compared with the present routes. Success by the optimization process was around 4-59 % for distance and 14-65 % for time. Consequently, a route optimization process on the street stationary container collection system will contribute a benefit by 24 % in total cost.

Full text

Global NEST Journal, Vol 9, No 1, pp 6-11, 2007
Copyright© 2007 Global NEST
Printed in Greece. All rights reserved
ROUTE OPTIMIZATION FOR SOLID WASTE COLLECTION:
TRABZON (TURKEY) CASE STUDY
O. APAYDIN* Yildiz Technical University
M. T. GONULLU Environmental Engineering Department
34349 Besiktas, Istanbul, Turkey
Received: 20/12/05 *to whom all correspondence should be addressed:
Accepted: 29/03/06 e-mail: apaydin@yildiz.edu.tr
ABSTRACT
In a municipal solid waste management system, decreasing collection/hauling costs, which
consist of 85 % of total disposal expenditure, can be carried out by a route optimization. Thus,
a huge amount of economical benefits is getting furnished. If route optimization is performed
in solid waste collection/hauling process, due to reductions in “empty miles” negativity, total
expenditures will be decreased.
Trabzon City located in the northeast side of Turkey has about 185 thousand inhabitants
according to Census 2000. The city shares just about 1% of the Gross Domestic Income in
Turkey. In other words, that means that Trabzon City livings have moderate revenue.
The objectives of this study are to optimize for the route of collection/hauling in Trabzon City
by taking consideration of data about road net, demographics and solid waste production.
In order to analyse the solid waste collection/hauling process in the city, the processes were
recorded by a Sony DCR-TRV145E brand video camera. To use route optimization process,
data related in present spending, truck type and capacity, solid waste production, number of
inhabitants and Global Positioning System (GPS) receiver data for each route were collected
and all the data were analyzed with each other.
For 39 districts in the city, a shortest path model was used in order to optimize solid waste
collection/hauling processes, as minimum cost was aimed. The Route View ProTM software as
an optimization tool was used for that purpose. Geographic Information System (GIS)
elements such as numerical pathways, demographic distribution, container distribution and
solid waste production amount were integrated to the software. To give an idea, thematic
container layer has 777 container location points for the entire city.
After performing routes by the software, the optimized routes were compared with the present
routes. Success by the optimization process was around 4-59 % for distance and 14-65 % for
time. Consequently, a route optimization process on the street stationary container collection
system will contribute a benefit by 24 % in total cost.
KEYWORDS: GIS, Trabzon City, route optimization, solid waste, collection/hauling.
1. INTRODUCTION
Municipal solid waste collection (MSWC) has about 85% proportion of the total cost for solid
waste management system [1]. MSWC is the beginning of the process of solid waste
management which consists of generation, collection, transfer, treatment and final disposal.
Integrated solid waste management involves a variety of programs and facilities, and
incorporates source reduction, reuse, recycling, composting, incineration and landfilling.
Waste stream from a city to any destination is charged a unit hauls cost based on per-ton
distance. However, waste stream of rejects from a processing facility to conversion or
disposal facility is ignored, because it has no significant effect. Typical haul costs are in the
range from 0.07 to 0.21 US$ km-1 ton-1 for collection vehicles, while transporting waste by
transfer trailers reduces costs to 0.03 to 0.10 US$ km-1 ton-1 [2]. Population of Turkey is
67,844,903 according to Census 2000 [3]. Turkey takes place in the group of medium income

ROUTE OPTIMAZION FOR SOLID WASTE COLLECTION
7
level. There are 3215 municipalities in Turkey. Only 11 municipalities had a sanitary landfill in
2000. 12 % of the total population of Turkey lives in cities located in Black Sea Region.
Trabzon City is one of cities in the region. In the region, Sanitary Landfill does not exist. It is
determined that MSWC unit cost is between 0.04-0.06 US$ km-1 ton-1 and haul cost is
between 0.02-0.04 US$ km-1 ton-1 [4]. Annual income of inhabitants living in Trabzon City is
US$6100 (min: 1800, mean: 6100, max: 42000) [5]. A study for Istanbul emphasizes that
yearly collection expenses of solid waste can be reduced about 50% when an optimization
effort is used [6]. According to another study, for developing countries, ratio of total collection
costs was determined to be approximately 79% for low income, 74-79% for middle income
and 55-70% for high income [7]. Cost, of course, is proportional to distance. Conventional
dispatching methods have generally focused on minimizing. Decision support system
approach, a dispatching problem, has a focus on empty miles minimization [8]. Studies
performed in a small district of Trabzon city, by route optimization, pointed out a success of
22 % reduction for collection time and 20 % reduction for collection distance for the MSW
collection processes [9, 10]. It is necessary that empty miles minimization is performed on
MSW collection/hauling processes.
Present paper’s objectives are:
1. To create MSWC data recorded by a video camera by riding in collection vehicle
cabined.
2. To use data gathered by observing the collection of MSWC in 39 districts in the city
and to form a GIS database on the route map of Trabzon
3. To compare present route with optimized route for cost and time
2. SOLID WASTE MANAGEMENT IN TRABZON CITY
To collect wastes, Trabzon municipality has located about 2,800 stationary containers in
different sizes (150, 300, and 400 L) in the residential area. Wastes from households are
dropped into these containers by inhabitants. Waste containers are unloaded at least twice a
week by 20 trucks that have a total capacity of 154 m3. Total length of road network is about
416 km and collection vehicles travel 60 % of that distance each day. On the other hand,
collection through some central busy streets such as Maras Caddesi is performed 7 or 8
times a day. The number of total daily tour reaches at 50 for MSWC process in the city.
MSWC facility is subjected for 6 days a week. The city has no transfer station, yet. Collected
garbage is dumped at the seaside of Black Sea by being blended with demolition waste and
soil in a ratio of about 50%. The dumping area with 2 Ha has been prepared as surrounded
by breakwater walls in the sea [4]. Annual income of inhabitants living in Trabzon City is 6100
US$ (min: 1800, mean: 6100, max: 42000) [5].
To begin understanding collection process in the city, some daily operational data was
determined for collection vehicles and presented in Table 1 and Table 2.
Table 1. Route number and total collected MSW subjected for each truck per day
Vehicle registration
plate ID Vehicle route number,
route d-1 Total MSW amount
collected, kg d-1
61 DE 762 4.17 11571
61 DE 761 3.67 9630
61 DE 772 3.67 9736
61 DE 773 3.67 7417
61 DE 775 3.67 9615
61 DE 766 3.5 8354
61 DE 771 2.5 7295
61 DE 763 3.67 11171
61 DU 367 4 9576
61 AE 398 3 6456
Total 35.5 90823
Mean 3.56 9082.3
Standard deviation 0.48 1673.7

APAYDIN et al.
8
Table 2. Densities of MSW collected in trucks in work days
MSW density in collection truck,
kg m-3
Work days Mean Standard
deviation
Friday (1st Week) 343 114
Saturday 377 163
Monday 387 94
Tuesday 357 106
Wednesday 392 106
Thursday 344 119
Friday (2nd Week) 385 101
Mean 369 115
Standard deviation 21 23
3. MATERIAL AND METHOD
In this study, a video camera, Sony DCR-TRV145E, was used in order to analyse solid waste
collection/hauling process. By this way, more realistic solid waste collection/hauling costs
were produced based on data obtained from records. All data were stored in a GIS database.
Projection of digitizing map used in this study has been adjusted Turkish Coordinate System
((GK 3 Degree k=1-ED50 and Category Members: GK Central Meridian 39 (ED50)). The map
was containing several layers related in 39 districts. A shortest path model was used in order
to optimize solid waste collection/hauling processes, by aiming at minimum distance. The
Route View ProTM software integrated with GIS elements such as numerical pathways,
demographic distribution, container distribution and solid waste production was used as an
optimization tool.
3.1 Determining capacities of containers and vehicles
Container number (nK) is computed as follows
KK
p
nP
=,
where p is population living in the area and PK is population for a container. PK is written as
K
KRP
V
PV
=,
where VK is the volume of a container (m3) and VRP is the volume of MSW per person (m3).
VRP is found this way:
R
RP R
V
VP
= or RP p
M
VW
=,
VR is the volume of MSW per residence (m3) (VR = 4.1 VRP), PR is the number of people per
residence, M is MSW amount per person a day (kg/person.day), Wp is unit volume per kg of
MSW in a container. MSW container number to be collected by a vehicle ( '
K
n) is presented
as follows
'v
KK
V
nV
=α
,
where Vv is volume of a vehicle (m3) and
α
is vehicle compaction factor.
The map performed in this study is presented in Figure 1. There are a lot of data having been
presented in the figure. Those are:
• Population of study area (Trabzon Municipality is located in The Eastern Black Sea
Region of Turkey) is 223976
• The number of districts that exist in Trabzon Municipality are 39
• Total area of the study area is 28689576 m2
• Total road distance traveled by collection vehicle is 416174 m
• The Number of residences living in Trabzon Municipality is 58909

ROUTE OPTIMAZION FOR SOLID WASTE COLLECTION
9
• Total number of containers used this study are 777 (the volume of each container is
0.8 m3)
• Temporarily solid waste disposal area is located in the north of Trabzon City)
Figure 1. The map of Trabzon City
4. RESULTS
Routes optimized by using the software were compared with present routes. The comparison
results are presented in Table 3. According to the Table, if the optimized routes are used in
solid waste collection system, both distance and time will be decreased by 4-59 % and 14-65
%, respectively.
After route optimization in the city, optimized total route numbers and total collection and
hauling travel distances for truck type/types per day obtained by using present vehicles and
containers are given in Table 4. Moreover, Table 5 presents optimized collection and hauling
costs obtained from data for possible usage of truck type/types. Total distance and costs of
collection/hauling for optimized case of present infrastructure are determined to be 366 km
and US$1844 per day. Monthly costs for optimized and present route are US$55320 and
US$73334, respectively. This difference expresses a decrease by 24.7 % in cost. Table 6
illustrates a cost comparison matrix obtained by binary comparisons of vehicle types. From
Table 6, it is seen that truck in 15 m3 will be the most economical capacity to collect all
containers in the city.
5. CONCLUSIONS
In solid waste management system, collection of solid waste is the most important process for
total disposal costs. In order to decrease total solid waste disposal costs It is necessary to
performed route optimization on current solid waste collection paths. This optimization study
supported data by video camera from field puts forward that the optimization process supplies
successes 24.7% in distance and 44.3% in time for collection and hauling. Accordingly,
24.7% benefit in total expenditure will be acquired. Furthermore, some extra benefits such as

APAYDIN et al.
10
exhaust and noise emissions, traffic jam, resource saving, etc., which are possibly more
important for city life quality than cost, will be acquired by the route optimization
Table 3. Time and distance comparisons for present and optimized cases of routes
Present routes Optimized routes Advantage, %
MSW
collection
route name Route
distance, m Route
time, s Route
distance, m Route
time, s Distance Time
Route 1 4833 2150 1966 702 59 67
Route 2 2590 1498 1481 529 43 65
Route 3 2997 998 2574 859 14 14
Route 4 6930 4016 6149 2202 11 45
Route 5 2621 1477 2523 900 4 39
Route 6 2751 1580 2015 668 27 58
Route 7 3337 1491 3090 1104 7 26
Route 8 3034 1760 1801 644 41 63
Route 9 5577 2148 4696 1682 16 22
Mean 3852,2 1902 2921,7 1032,2 24,7 44,3
Table 4. Optimized route numbers and travel distances for truck types
Vehicle travel distance
per day, km
Vehicle capacity Total
container
number, nK
Container
number per
vehicle
(
α
=2), '
K
n
Vehicle
route
number
per day collection hauling
7 m3 777 17 46 138 276
12 m3 777 27 29 145 174
15 m3 777 37 21 147 126
All of the vehicles 777 - 36 150 216
Table 5. Costs developed for optimized collection/hauling as depending on truck types
Costs, US$ day-1
Vehicle capacity collection
(0.05 US$ km-1 ton-1)hauling
(0.02 US$ km-1 ton-1) Total
7 m3 1077 861 1938
12 m3 1131 543 1674
15 m3 1147 394 1541
All of the vehicles 1170 674 1844
Table 6. A matrix developed by binary comparison of vehicle collection costs
% Benefit =
(Column/Line) *100 7 m3 12 m3 15 m3 All of the
vehicles
7 m3 0.0 15,7 25,7 5,1
12 m3 -15,7 0.0 8,6 -10,1
15 m3 -25,7 -8,6 0.0 -19,6
All of the vehicles -5,1 10,1 19,6 0.0

ROUTE OPTIMAZION FOR SOLID WASTE COLLECTION
11
ACKNOWLEDGEMENTS
The authors thank the Basarsoft for their assistance in the software used in this study.
Furthermore, they thank the Municipality of Trabzon. This research was supported by Yildiz
Technical University Scientific Projects Coordination Department. Project Number: 25/05/02.02.
REFERENCES
1. Ludwig H.F. and Black R.J. (1968) Report on the solid waste problem. Journal of Sanitary
Engineering Div., 94(2), 355-370.
2. Modak A., Everett J.S. (1996) Optimal regional scheduling of solid waste system. II: Model
solution, ASCE Journal of Environmental Engineering, 122(9), 793-799.
3. SIS (2002) Population Censuses, year 2000: City 61-Trabzon, ISBN 975-19-3065.
4. Apaydin, O., Geographical information system (GIS) supported optimization of solid waste
management alternatives in Trabzon City, Ph.D. Thesis, Yildiz Technical University, Science
Institute, Environmental Engineering, Istanbul, 2004.
5. Apaydin O., Kalender A., Gonullu M.T. (2002) Assessment of Sociological Aspect of Separate
Household Solid Waste Collection in Trabzon (Turkey), ISWA 2002 World Environmental
Congress and Exhibition, 8-12 July 2002, Istanbul, Turkey.
6. Kinaci C., Gorgun E., Arslan M., at. al. (2000) Private Sector Participation in Municipal Solid
Waste Services-A Case Study for Kadikoy of Istanbul in Turkey, Wastecon 2000, Biennial
Conference and Exhibition on Integrated Waste Management in The Millennium, 5-7
September 2000, Somerset west near Cape Town, South Africa.
7. Cointreau-Levine, S.C., (1994), Private Sector Participation in MSW Services in Developing
Countries, Volume 1. The Formal Sector, The World Bank, Washington D.C., USA.
8. Taylor G.D., Whicker G.L., Usher J.S. (2001) Multi-zone dispatching in truckload trucking,
Transportation research Part E, 37, 375-390.
9. Apaydin, O., (2005) A GIS Supported Optimization Application of Solid Waste Collection
System in Trabzon City, Ekoloji, 54, 1-6.
10. Apaydin, O., Arslankaya, E., Avsar, Y., Gonullu, M.T., (2004) GIS supported optimization of
solid waste collection: Trabzon case study, Sigma, 4, 249-254.