Advances in Engineering Research (AER), volume 147

Conference on Global Research on Sustainable Transport (GROST 2017)

OPTIMIZING DELIVERY ROUTE BY APPLYING

MILKRUN METHOD
Putri Setiani1, Hanifah Fiddieny2, Edhie Budi Setiawan3, Diah Etika
Cahyanti4
1. STMT Trisakti, 2. STMT Trisakti, 3.STMT Trisakti, 4. STMT Trisakti
corresponding author:[email protected]

Abstract: Efficiency of the optimal vehicle route to generate effective delivery is an

important issue in the logistics world. The common methods usually practiced in
delivery systems are direct shipment, indirect shipment and milk run method. The
objective of this research is to analyze the optimization of delivery route by
comparing direct shipment method to milk run method to obtain effective delivery.
This research involved 8 suppliers and 1 delivery route from the total population of
18 suppliers at PT. Toyota Motor Manfacturing Indonesia and 4 delivery routes in
Cikarang area. Data were analyzed by using Vehicle Routing Problem and saving
matrix method. The result shows that the optimal delivery route can minimize the
distance and vehicle leadtime, therefore, the total cost spent in the milkrun delivery
process is cheaper than by applying direct shipment method. It can be concluded
that regular parts shipment process by using milk run method is more effective than
by using direct shipment method.
Keywords: Optimal Route, Milk Run, Direct Shipment

Introduction
Delivery system plays an important part in logistics. Delivery of goods
in a country is dependent to public infrastructure of that country. Then the
delivery process needs a high efficiency in logistics activities, including the
level of security quality, safety and service (S, 2016). The common delivery
system applied in most companies in distributing goods is the direct shipment
method. According Li, Wu, Lai, & Liu, (2008) direct shipment method is a
delivery method that is practiced in a company in which the supplier provides
a vehicle used per period of time to one customer or a retailer. Another
definition explained by Sadjadi, Jafari, & Amini (2009), direct shipment is a
direct delivery from a supplier to its customer. In other words, direct
shipment is a delivery system done directly from a supplier to a customer by
using a vehicle to serve one specific customer.

Copyright © 2018, the Authors. Published by Atlantis Press. 748

This is an open access article under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).
 Advances in Engineering Research (AER), volume 147

Table 1 Total Cost of Direct Shipment

DIRECT SHIPMENT Distance/10 Shipment

Total Cost/10 Days
ROUTE days Volume
A-DC-A 460,00 1,12 Rp 10.060.000,00
B-DC-B 494,80 0,33 Rp 10.442.800,00
E-DC-E 485,20 1,73 Rp 10.337.200,00
F-DC-F 467,20 12,2 Rp 14.139.200,00
G-DC-G 484,80 0,46 Rp 10.332.800,00
H-DC-H 475,20 4,24 Rp 10.227.200,00
J-DC-J 482,80 5,67 Rp 10.310.800,00
L-DC-L 461,60 10,91 Rp 14.077.600,00
TOTAL 3811,60 36,66 Rp 89.927.600,00
The increase number of production has obliged PT. Toyota Motor
Manufacturing Indonesia to reconsider about the route and the delivery cost.
A study done byBell & McMullen (2004), confirmed that efficiency of the
vehicle route is an important issue in logistics business, as the optimal
delivery route will generate effective delivery. This is also supported by Sari
(2015) efficiency and effectiveness in the delivery process needs to know the
best route. One way to enhance efficiency in delivering goods is by applying
the Milk run method. Milk run method is a delivery system in which vehicles
are rotating based on a schedule defined before, stopping at every department
to collect products that have been ordered (Kovács, 2011).
Derived from the condition above, this research is conducted to find an
optimal route for milk run delivery method in achieving the effective regular
parts shipments. How effective is the milk run delivery method compared to
direct shipment method?
The research is done to find out the optimal delivery by practicing Milk
Run system to achieve the effective and efficient delivery system. Nemoto
& Rothengatter (2012) stated that the Milk run delivery system deals with
suppliers of a company or several suppliers from different cooperating
companies and the delivery is done by the third party logistic (3PL) or fourth
party logistic (4PL). Whereas in the process of material handling, milk run
performs as the waste and cost reduction under consideration that this system
does the delivery by using one mobile vehicle periodically in a certain route

749
 Advances in Engineering Research (AER), volume 147

so that it will reduce fuel consumption (Kilic & Durmusoglu, 2012).

Furthermore, the result of the milk run system is reducing transportation cost,
route and fuel consumption (Kovács, 2011). Based on the previous
researches, it can be concluded that milk run delivery system is a system that
uses one vehicle rotating and visiting every customers and suppliers in one
route based on order. It is done to increase efficiency of the vehicle (truck),
to reduce delivery cost and to achieve effective delivery.
According to Sadjadi et al., (2009) direct shipment method is where the
consignments are directly delivered to the manufacturers from the suppliers,
and the company is responsible for storing the general inventory. This logistic
pattern is traditionally used for supplying all parts. In other words the
shipment is done directly from supplier to customers, the advantage of
applying this method is the goods can be delivered faster without having to
wait for another customers. However it takes more vehicles and more cost to
spend.
A research done by Iqbal (2013) suggested that optimization is a way in
finding optimal solution, at the same time, overcoming resource constraints
and minimizing cost, maximizing profits and improving process efficiency.
Witlox (2007) stated, delivery route planning involves ordering center in a
certain area, so that it can optimize traveling time, and transport operation
between customer and supplier. A company that is able to cut down the
delivery route and lessen the number ofvehicles used would be able to
increase its added value.In accordance with a research done by Endarwati,
Mahmudy, & Ratnawati (2014), to shorten delivery process time, companies
should choose the most optimal routes. By applying the milk run method, it
is expected to reduce the costs during delivery process, reduce the use of
warehouse space, and improve the supply chain performance or logistic as a
consequence of the effective transport process (Sadjadi et al., 2009).

750
 Advances in Engineering Research (AER), volume 147

Method
In this research, researchers used delivery volume data and delivery
route data from several suppliers to PT. TMMIN. Data were gathered by
doing observation in logistic division of PT. TMMIN in September to
November 2017. Researchers took samples from 8 suppliers and 1 delivery
route from the total population of 18 suppliers and 4 delivery route in
Cikarang area. In analyzing data, researchers employed the Vehicle Routing
Problem (VRP) method that is applied to decide the optimal route which will
be taken by the vehicles, focusing on one depot or more to serve several
customers. There are some conditions in the application of vehicle routing
problem, for instance, each route should not exceed vehicle capacity, and the
total length of each route should not be greater than the set limit. The general
purpose of VRP are to minimize distance and costs of the vehicles, minimize
the number of vehicles used to serve all costumers, balance time per route,
and so forth (Toth & Vigo 2002). A research done by Poot, Kant, &
Wagelmans (2002) explained that saving method is a method to minimize
distance, time and costs by considering the constraints. Coordinate points of
the original distance to the destination will be measured, and the millage of
all vehicles will be minimized afterward.
This research also employed steps in saving matrix method initiated by
Pujawan&Mahendrawathi (2010). The first step is determining the
coordinates of each supplier locations to recognize distance between
locations. The second step is calculating distance saving by using saving
matrix. Integrating two or more suppliers into one delivery route. The third
step is allocating suppliers into vehicle or delivery route. Truck specification
from 3PL party which will transport the regular parts from every supplier in
one route is wing box trucks, the specifications are as followed:
Table 2 Transport Capacity Wing Box Truck
Max Wing Box Truck Capacity 7.5 x 2.4 x 2.4 43.2 m3
Buffer for Loading Operation 0.2 x 7.5 x 2.4 3.6 m3

Transport Capacity 39.6 m3

751
 Advances in Engineering Research (AER), volume 147

The fourth step is determining truck visit order to each supplier by

sweep method. Determining the order of trucks visit started from DC then
found the nearest supplier from DC, proceeded to the next suppliers clock-
wise or counter clock-wise, additional suppliers in each visit would be
accumulated based on its distance, hence the length of the distance that must
be taken by truck will be obtained. Another explanation was given by
Supriyadi, Mawardi, & Nalhadi (2017) saving method with sweep method is
used to decide vehicle schedulein optimal route. Saving matrix is also applied
to determine products distribution route to target market by deciding the
distribution route that should be taken and the number of vehicles which is
suitable with the capacity to attain the shortest route and minimum costs. The
visit order would then be determined by sweep method. Sweep is a part of
saving matrix method used to decide the order of visit routes, done by drawing
a straight line from DC to the nearest destination then proceed to another
destination clock-wise or counter clock-wise. Hourilson, Souza, &
Fluminense (2011) also explained that the sweep method is a method done as
the first step of the movement in determining order of visit to several
suppliers. The result of the optimizing milk run method from the four steps
above will then be compared based on costs, distances, and leadtime to the
one which is done in direct shipment method.

Discussion and Result

A. Route Analysis
The steps in determining optimal route by saving matrix derived from
Pujawan&Mahendrawathi (2010) are:
1. Determining the DC coordinate with suppliers’ location and
coordinates among suppliers.
Table 3 Coordinates of suppliers location and the amount of delivery volume.

Suppliers’ Location Delivery Coordinates

No Suppliers
code (Cikarang) Volume (m3) X Y
Sari Takagi
1 A Jababeka 12,2 9,9 -6,2
ElokProduk

752
 Advances in Engineering Research (AER), volume 147

Table 3, Cont.
Coordinates of suppliers location and the amount of delivery volume.

Takagi Sari Delta

2 B 0,46 10,1 -6,7
Multi Utama Silicon
3 E Otics Indonesia EJIP 10,91 8,85 -7,4
Ingress Malindo
4 F Jababeka 2 1,73 10,4 -6,25
Ventures
Kojima Auto
5 G Mall Lippo 4,24 9,45 -7,2
Technology
6 H MTAT Indonesia Jababeka 1 1,12 10,15 -5,4
7 J Sanoh Indonesia Hyundai 5,67 9,25 -7,75
Citra
8 L Hyundai 0,33 11,2 -5,25
NugerahKarya
As seen in table 1.3 the coordinates of each location, in this research is
the distance in points (X, Y) describes the actual distance that should be taken
by the vehicle and the length of time taken by the vehicle. The identified
coordinates are put in saving matrix table.

Table 4 Distance between suppliers and Distance from DC to each suppliers

DC A B C D E F G H I J K

A 11,7 0,0
B 12,1 0,5 0,0
C 11,1 1,3 1,4 0,0
D 11,2 1,6 1,6 0,3 0,0
E 11,5 1,6 1,4 0,6 0,4 0,0
F 12,1 0,5 0,5 1,8 2,0 1,9 0,0
G 11,9 1,1 0,8 0,8 0,9 0,6 1,3 0,0
H 11,5 0,8 1,3 2,0 2,3 2,4 0,9 1,9 0,0
I 11,0 0,8 1,3 1,5 1,8 2,0 1,2 1,7 0,7 0,0
J 12,1 1,7 1,4 1,1 0,9 0,5 1,9 0,6 2,5 2,3 0,0
K 11,4 0,9 0,9 0,5 0,7 0,8 1,3 0,6 1,6 1,2 1,1 0,0
L 12,4 1,6 1,8 2,9 3,2 3,2 1,3 2,6 1,1 1,7 3,2 2,5

753
 Advances in Engineering Research (AER), volume 147

2. Calculating the saving distance to find the route combination.

Table 5 Saving distance among suppliers points

DC A B C D E F G H I J K
A 11,7 0,0
B 12,1 23,3 0,0
C 11,1 21,4 21,8 0,0
D 11,2 21,3 21,7 21,9 0,0
E 11,5 21,6 22,2 22,0 22,3 0,0
F 12,1 23,3 23,7 21,4 21,3 21,7 0,0
G 11,9 22,5 23,2 22,1 22,2 22,8 22,7 0,0
H 11,5 22,3 22,3 20,5 20,4 20,7 22,7 21,5 0,0
I 11,0 21,9 21,8 20,5 20,3 20,5 22,0 21,2 21,8 0,0
J 12,1 22,1 22,8 22,0 22,3 23,1 22,3 23,4 21,1 20,8 0,0
K 11,4 22,2 22,6 21,9 21,8 22,1 22,2 22,7 21,3 21,1 22,4 0,0
L 12,4 22,4 22,7 20,5 20,3 20,7 23,2 21,6 22,8 21,7 21,3 21,3
The integration is started from the biggest savings values in order to
maximize savings. Two or more suppliers are integrated in one route. Table
1.5 shows that the biggest distance savings in the route is in suppliers B, F,
G, J, A, L, E, and H with the total delivery volume in this route is 36.66 m 3,
the delivery volume is suitable to the capacity of the truck. Next is counting
the efficiency gained from regular transport by using wing box
36,66
trucks: 39,6 x100% = 93%

3. Determining truck visit order by Sweep method

From the calculation of truck visit order to each supplier, the total
millage obtained equals to 163.26 cm:

Table 6 truck visit order by SweepMethod

DC H L F A B G J DC DISTANCE
11,50 0,81 23,22 23,31 23,26 23,18 23,36 23,08 163,26

B. The calculation of Distance, Time, and Costs

In this phase, the researchers calculate the total distance, lead time and
total costs of milk run delivery method in trial period, starting from

754
 Advances in Engineering Research (AER), volume 147

September 11, 2017 to September 24, 2017. Therefore, the effective work
days in trial period are 10 days. Distance on the map shows 1.5 cm to every
2 km of actual distance. (Cost of renting a wing box truck/day is Rp.
1,500,000 and cost/km is Rp. 11,000).

Table 7 the result of calculation of actual distance, leadtime and total costs

Loading
Actual Actual Delivery Total
unloading
Distance/ Distance/10 time delivery Total Costs
time
day (km) days (km) (minutes) time
(minutes)
363,18
326,52 3265,20 36,66 326,52 minutes = Rp50.917.200
6,05 Hour
By assuming that the standard of loading/unloading time per m 3 is 1
minute and standard vehicle speed is 60 km/hour.

Table 8. Comparison of milkrun method to direct shipment method.

Actual Total Difference
Delivery Difference Difference Total Cost
distance/10 Delivery in Total
Method in distance in time (Rp)
days (km) time Cost
363,18
minutes
Milkrun 3265,2 50.917.200
= 6,05
Hour
14% 13% 43%
417,82
Direct minutes
3811,6 89.927.600
Shipment = 6,96
Hour

Based on the calculation above, it describes that the difference of the

millage in direct shipment method and the milk run method is 14% and the
reduction of vehicle leadtime is 13%while the costs is in 43%

Conclusion
The result of this research shows that the application of Milk run method
is more efficient and more effective compared to the direct shipment method.
The milk run delivery method enables to minimize the delivery costs, improve
distance efficiency, and minimize leadtime by finding the optimal route to get
the best route combination by integrating several suppliers into one route
within the capacity of trucks’ total delivery volume.

755
 Advances in Engineering Research (AER), volume 147

Suggestion, in order to attain more optimal result, further research concerning

to the other delivery route needs to be conducted.

References
Bell, J. E., & McMullen, P. R. (2004). Ant colony optimization techniques
for the vehicle routing problem. Advanced Engineering Informatics,
18(1), 41–48. https://doi.org/10.1016/j.aei.2004.07.001
Endarwati, D. A., Mahmudy, W. F., & Ratnawati, D. E. (2014). Pencarian
Rute Optimum Dengan Evolution Strategies. DORO: Repository Jurnal
Mahasiswa PTIIK Universitas Brawijaya, 4(10). Retrieved from
http://filkom.ub.ac.id/doro/
Iqbal, K. (2013). Fundamental Engineering Optimization Methods.
Jamilson, M., Souza, F., & Fluminense, U. F. (2011). A hybrid heuristic ,
based on Iterated Local Search and GENIUS , for the Vehicle Routing
Problem with Simultaneous Pickup and Delivery Marcio Tadayuki Mine
, Matheus de Souza Alves Silva , Luiz Satoru Ochi and Anand
Subramanian, 10(2), 142–157.
Kilic, H. S., & Durmusoglu, M. B. (2012). Classification and modeling for
in-plant milk-run distribution systems, 1135–1146.
https://doi.org/10.1007/s00170-011-3875-4
Kovács, A. (2011). Optimizing the storage assignment in a warehouse served
by milkrun logistics. International Journal of Production Economics,
133(1), 312–318. https://doi.org/10.1016/j.ijpe.2009.10.028
Li, J. A., Wu, Y., Lai, K. K., & Liu, K. (2008). Replenishment routing
problems between a single supplier and multiple retailers with direct
delivery. European Journal of Operational Research, 190(2), 412–420.
https://doi.org/10.1016/j.ejor.2007.06.022
Nemoto, T., & Rothengatter, W. (2012). Efficient Green Logistics in Urban
Areas: Milk Run Logistics in the Automotive Industry. Sustainable
Transport for Chinese Cities (Vol. 3). Emerald Group Publishing
Limited. https://doi.org/10.1108/S2044-9941(2012)0000003017
Poot, a, Kant, G., & Wagelmans, a P. M. (2002). A savings based method
for real-life vehicle routing problems. Journal of the Operational
Research Society, 53(1), 57–68.
https://doi.org/10.1057/palgrave/jors/2601252
Pujawan&Mahendrawathi, institute teknologi sepuluh N. (2010). Supply
CHain Management (2nd ed.). Guna Widya.
Raden Didiet Rachmat Simarmata, J. H., & Len Togas, N. (2015). Study on
The Cause of Stock Out of Logistic Installation in Harapan Kita
Hospital. Jurnal Manajemen Transportasi & Logistik, 2(1), 79–92.
S, T. I. H. (2016). MULTIMODA DALAM MEWUJUDKAN VISI
LOGISTIK INDONESIA 2025, 69–84.
Sadjadi, S. J., Jafari, M., & Amini, T. (2009). A new mathematical modeling
and a genetic algorithm search for milk run problem (an auto industry
supply chain case study). International Journal of Advanced

756
 Advances in Engineering Research (AER), volume 147

Manufacturing Technology, 44(1–2), 194–200.

https://doi.org/10.1007/s00170-008-1648-5
Sari, L. M. (2015). PENGIRIMAN PRODUK ARNOTTS KE ALFAMART
AREA JABODETABEK DAN BANTEN PT INTAN UTAMA
LOGISTIK TAHUN 2015, (3), 11–20.
Supriyadi, S., Mawardi, K., & Nalhadi, A. (2017). Minimasi Biaya Dalam
Penentuan Rute Distribusi Produk Minuman Menggunakan Metode
Savings Matrix. Seminar Nasional Institut Supply Chain Dan Logistik
Indonesia (ISLI), (September), 1–7. https://doi.org/DOI
10.17605/OSF.IO/2D9GP
Toth, P., & Vigo, D. (2002). Models, relaxations and exact approaches for the
capacitated vehicle routing problem. Discrete Applied Mathematics,
123(1–3), 487–512. https://doi.org/10.1016/S0166-218X(01)00351-1
Witlox, F. (2007). Logistics systems for sustainable cities. Cities, 24(6), 462–
463. https://doi.org/10.1016/j.cities.2007.07.003.

757