Optimizing Sustainable Transportation Process Unloading Heavy

Equipment Company
Fikri Haekal Ibrahim 1, Dendhy Indra Wijaya 1*, and Hubertus Davy Yulianto1
1Industrial Engineering Department, BINUS Online, Bina Nusantara University, Jakarta, Indonesia 11480

Abstract. PT. XYZ, a heavy equipment company, faced a critical challenge during the months of May,
June, and July, specifically on June, when approximately 18 trucks entered its premises. Mismanagement of
this influx led to the formation of truck queues, highlighting the urgent need for effective handling. In
response to this challenge, our research aimed to determine the maximum capacity of PT. XYZ's facility
and implement strategies for optimizing the truck unloading process. The primary objective was to reduce
unloading times and waiting queues, ultimately minimizing the operational impacts. We employed the
single-channel, single-phase queuing method, facilitated by Arena software, to analyze capacity and
optimize queue times. Our manual calculations, using the single-channel, single-phase queuing method,
indicated an average queue time of 1.8 hours, with two trucks unable to be accommodated. The cumulative
queue time over three shifts amounted to 3.6 hours, totaling 21.5 hours. In contrast, the Arena software
simulation showed that 16 out of 18 trucks could be efficiently served, resulting in a queue time of 4.4 hours.
We proposed improvement scenarios based on total cost analysis within the Arena software. Scenario 2
demonstrated superior performance and cost efficiency compared to Scenario 1.

1 Introduction
PT. XYZ is a heavy equipment company that produces
heavy equipment components such as dozers, tractors,
loaders, backhoes, and excavators. PT. XYZ stores raw
materials with large loads in the warehouse or storage,
so it is necessary to unload the bar material by
considering the capacity. The forklift machine used to
unload the cargo from the trailer truck is operated by one
operator, who is divided into three working times
(shifts). This causes the company to face challenges in
the truck unloading process to avoid the buildup of truck Fig. 1. Graphic data plan received container/trailer
queues if the trailer truck stays in the process of
unloading material in storage. Queuing theory is a It can be seen in Fig. 1., a graphic data plan for
theory related to the mathematical study of waiting receiving containers from May to July. Based on June,
lines. Queuing theory is always related to all aspects of there is data on receiving containers and trailers. The
the customer situation to obtain a service. Queues are receiving container/trailer plan data does not yet know
people or goods in a line waiting to be served or the maximum capacity in one day that could be
covering parts of the company, which could determine accommodated by PT. XYZ. This is due to trucks that
the best time and facilities to serve customers efficiently are still in the queue for loading and unloading, so
[1]. It could be imagined that if the customer takes a long further evaluation is needed to optimize the duration of
enough waiting time, a small percentage of idle time will the unloading process with existing conditions.
be obtained, which means there is absolutely no idle This research is intended to be conducted to know the
time in the service [2]. Based on the data from the PT. maximum capacity at PT. XYZ and to improve and
XYZ container and trailer receive plan, including optimize the unloading process to reduce unloading time
companies that are always congested in the unloading in the waiting time queue, which could minimize the
process in storage, the company is always faced with a impact caused by optimizing and improving the process
queue of waiting trucks. If the truck stays overnight, and provide recommendations for process
there will be an overnight charge for the truck. improvements to achieve the optimal duration of
unloading at PT. XYZ to reduce the time required to be
more efficient and effective using simulation.

* Corresponding author: [email protected]

2 Methods 3. Problem Formulation
The thesis topic is narrowed down by finding a
The method used in this study combines qualitative and problem statement, structuring the research
quantitative research methods in the field. This study systematically, and preventing it from expanding. This
focuses on data collection methods, testing, and problem study aims to analyze the current capacity conditions at
analysis using the queuing method for the calculation of PT. XYZ, particularly the one-day waiting queue, and to
truck arrivals: a flow chart for this research. evaluate the existing situation. Furthermore, it seeks to
provide recommendations for process improvements
aimed at achieving optimal unloading durations in the
waiting queue at PT. XYZ, with the ultimate goal of
reducing the time required to enhance efficiency and
effectiveness.
4. Data Collection
Data collection is very important in this research to
prevent invalid research results. Primary data includes
finding monthly data on trucks that enter PT. XYZ.
Secondary data consists of interviews with sources.
Primary data such as data driver reports to security, data
check unload priority, data check material, and data
unload material to storage.
5. Data Processing
From the data collected, fitting distribution data
After that, verification and validation data are collected
help using SPSS program and a scenario improvement
design is created using Arena simulation.
6. Solution
Given recommendations that could be applied by
companies related to the implementation of making
control systems. scenario improvement using
simulation.
7. Evaluation
The study evaluates the queues available at the company
so that trucks could be accommodated in accordance
with or related to the data of incoming trucks. Use
validation and verification from data processing if the
data is not valid.

3 Result and Analysis

PT. XYZ is a company engaged in producing heavy
equipment components. It is necessary to carry out the
process of unloading bar material by considering the
capacity of the unloading truck.

Fig. 2. Research Step Table 1. Working hours for receiving trucks

Working hour receiving truck
In Fig. 2., described flow process research from start shift Hr
the research until finished. Divided into several parts.
1. Field Study 1 8
The field study began with observations of the truck 2 6.5
arrival process. Interviews were conducted with several 3 7
resources such as the manager, operator, helper, and
security, as well as problems that occur in the field that total 21.5
could be used as thesis topics
2. Literature Study
A literature study is carried out for reference to cases Table 1, to inform that at PT. XYZ, are working a
and topics of similar problems as an accurate reference receiving truck shift with a 3-time shift. In existing the
for completing the thesis; these keywords include forklift machine used to unload cargo from the trailer
keywords such as queues using Arena simulation and truck is operated by 1 operator and 1 helper, which is
the single channel single phase queueing method. divided into 3 working times (shifts) with a total
working time of 21.5 hours.
 In the section (Method), drop-off takes time due to drop-
off unloading taking time due to the forklift operated by
Table 2. Trucks Data
only 1 operator divided into 3 shifts. Furthermore, in the
environment section (environment), the main cause is
the storage place, which is far from the unloading place
which takes time from the unloading place to.

Table 3. Countermeasure of problem causing queue

Table 3 explains the countermeasure to the problem

queue causing a stack of trucks at PT. XYZ. Could be
seen in the table factor because the operator does not
realize that the truck is coming even though the operator
Table 2 shows the time between truck arrivals. The checking the unloading application has been provided,
queuing system is a set of services, and entities, a rule so more attention is needed, and operators must be swift
that regulates the arrival of these entities and the service and responsive for material and method using Arena
process [3]. Queuing analysis is a form of probability simulation.
analysis which is defined as people or goods in line who
are queuing [4]. Single Phase Single Channel is a single- Table 4. Distribution fitting results with arena software
phase system that has only one line to enter the service Process Distribution expression
Time Between Normal NORM
system, then after getting the service directly out of the
Arrival (42.5, 29.5)
system. Calculations using the queuing method Single Driver Report to beta 1.5 + 4 * BETA
Channel Single Phase obtained the average time in the Security (1.38, 0.93)
system (Ws) =1/((µ-λ)) [5]. With a total load of up to 18, Operator check beta 6.5 + 6 * BETA
there will certainly be a queue or buildup. We need to unload priority (0.744, 0.93)
know the amount of capacity that could be Check Material triangular TRIA
accommodated at PT. XYZ for 1 day on June 19, 2022. (3.5, 5.17, 5.5)
Calculations using the queuing method single channel Unload Material to Weibull 59.5 + WEIB
single phase obtained the average time in the system Storage (13.6, 1.49)
(Ws) =1/((µ-λ)) = 0.03 x 60 = 1.8 hours per truck. Total:
1.8 hours x 2 trucks = 3.6 hours. to find out the amount Table 4 explains the results of distribution fitting. The
of capacity that could be accommodated through the input analyzer is a facility within Arena simulation
FIFO method assisted by Arena software to find out the software whose aim is to do the distribution fitting
capacity that could be accommodated by PT. XYZ. before entering the data that was obtained at the time of
field observation into the simulation model. The data
obtained at the time of observation is fitted first for its
statistical distribution to be included in the simulation
model [7]. Based on the results of distribution fitting
using Arena software with the help of an input analyser
and through the fitting of all processes, it is found that
the result of the distribution This distribution fit was
chosen based on the smallest error.

Fig. 3. Analysis fishbone diagram

Fig. 3., the factors causing the problem are described
in the cause-and-effect diagram (fishbone diagram). In Fig. 4. Simulation existing setup on Arena software
the phase of identifying sources of risk, a fishbone
diagram is needed to identify and organize the causes Based on Fig. 4., Simulation is an imitation of the
that might arise from specific factors and then to operation process of a real condition or system over time
separate the root causes [6]. The buildup of truck queues [8]. Building simulations could be done using
at PT. XYZ is analyzed from the human factor (man) programming languages, simulation languages and
because the operator does not realize that the truck is using simulation software [9]. Modules from different
coming even though the Operator Checking Unloading panels can be used to build the same model. Arena has
application has been provided. In the section (Material), a fully hierarchical structure to maintain modelling
the number of trucks increases due to peak time flexibility. Generally, users pull in low-level modules
conditions, and there are also concurrent truck arrivals. from the Blocks and Elements panel, so that simulation
language flexibility is achieved. The modules are treatment marked by the difference or not after
composed of SIMAN components in Arena [10]. treatment [15].
According to [11][12][13], the sequence of processes The paired sample T-Test test has conditions :
that must be carried out in creating a simulation model 1. The data is normally distributed.
is the cycle diagram simulation existing form at PT. 2. The data owned is an interval or ratio comparison.
XYZ using single channel single phase with Arena Hypothesis for this case :
software is condition existing form based on the H0 : The value results of the simulation output with
sequence passed by a truck starting from the time existing / real conditions are not significantly different
between arrivals, then the driver checks with the H1 : The results of the value of the simulation output
security officer, the operator checks unload priority, with existing / real conditions are significantly
checks the material, and finally unloads the truck to different
storage. The forklift machine used to unload cargo from The confidence interval value is 95% with an alpha
the trailer truck is operated by one operator and one value of 0.05 which means the error rate value is 5%.
helper. The value (sig 2-tailed) of 0.875>0.05 which indicates
that it fails to reject H0 / accept H0 reject H1. meaning
that there is no significant difference between the
simulated value of the Arena and the real output value.

Fig. 5. Verification from simulation process

Fig. 5, the verification process is a verifying process
that is useful to determine whether the simulation model
created has run as expected and to ensure that the
simulation model is free from errors and runs in
accordance with the expected concept [14]. This is a
check for simulations existing. table shows that No error
or warning on the model.
Fig. 7. Output result existing condition Arena Simulation
Fig. 7 explains that after the simulation that could be
obtained from the simulation that has been done, there
are 16 trucks with an average waiting time for unloading
trucks to storage using single channel single phase takes
4.4 hours with a utility rate value of 98%, indicating that
this process is very busy.
(a)

Fig. 8. The suggested form of simulation improvement

proposal scenario 1

Fig. 8 explains the proposed simulation

(b)
improvement scenario 1 for continuous improvement.
Fig. 6. Validation using normality test (a) and Validation With the addition of 1 helper and 1 forklift operator with
using paired sample T-test (b) forklift rental in the unloading section of the truck for
storage, the results of the calculation are divided by 2
Fig. 6, explains the normality test using SPSS (a). because they are added with 1 helper and 1 operator and
Testing is done to determine whether the data is then entered into the input analyser for data fitting again.
normally distributed or not. Because the data received The simulation form still uses a single channel and
is 18 and less than 30, we see the results of sig. single phase, but in the unloading section, the expression
Shapiro-Wilk. Shapiro Wilk, based on the sig output. is changed to 29.5 + WEIB (7.38, 1.66), after calculating
shapiro- wilk If the sig.>0.05 value means the data is the unloading time of the new truck. This approach uses
normally distributed, The numbers 0.699 and 0.730 a simulation of adding a total workforce of 2 people, so
are obtained, which means that the value of both is that the total workforce for now is 4 people, and the
greater than 0.05. which indicates that the data is addition of 1 forklift rental for a total of 2 forklifts.
normally distributed.
After we test using normality test we validated
using a paired sample T-test (b). Paired sample T-test
is a test that is used to examine the effectiveness of
 unloading trucks to storage of 0.3525 hours, or about 21
minutes, with a utility level value of 50%. transit area
with an additional capacity of 160 tons (2 shelves). The
transit area could reduce the truck drop-off time.

Table 5. Result all of Simulation

Fig. 9. The suggested form of simulation improvement No Simulation Waiting Utilization Truck Could
proposal scenario 2 Model Time be
Accommodated
Fig. 9 describes the proposed form of simulation 2 1. Model 4.4 hour 98% 16
for continuous improvement. The proposed form of existing
simulation with the addition of direct transit There is no 2. Improvement 13.2 50% 18
queue with a multi-phase, single-channel model. If the scenario 1 minute
server is "unloading trucks directly to storage" then the 3. Improvement 21 50% 18
truck will enter the server directly, unloading directly to scenario 2 minutes
transit". for the mode of decision or selection given the
build expression formula "unloading directly to storage. Cost analysis is to calculate how much it costs to
WIP <= unloading directly in transit. WIP," which aims work. The calculated service costs, waiting costs and
for trucks that enter unloading directly to storage there total costs are calculated in each scenario. The following
is a queue that could directly enter unloading transit. is the calculation of queuing costs. Cost of service (CS)
= (employee salary)/(number of working hours). cost of
waiting (CW) Calculation of service costs for the
addition.

Salary UMR Cikarang (UMK 2023) = Rp.5,100,000

- Total working hours = 21.5/3 = 7.2
- Truck rental price if overnight charge = Rp.
1,650,000
- Forklift rental per day = Rp. 7,000,000/day
- providing outdoor shelves = Rp. 2,000,000 /
piece
Calculation of service cost (CS) with 1 person
CS = (employee salary) / (number of working
hours)/number of working days)
CS = Rp. 35,416
(a)
1. Existing Simulation Conditions
Calculation of waiting costs (CW)
Trucks that couldn’t enter experience overnight
charges
Existing Simulation Model
*) rental fee is charged if the waiting time is above 3
hours.
= truck rental price x waiting time range

= Rp. 1,650,000 x 4.4

= Rp. 7,260,000
Total = E(CW) = CS + CW = (Rp. 35,416 x 2 = Rp.
70,832) + Rp. 7,260,000 = Rp. 7,330,832

(b) 2. Proposed Simulation Conditions

Fig. 10. Result output proposal improvement simulation 1 (a) a. Simulation of proposed truck unloading with
and Result output proposal improvement simulation 2 (b) additional operators and forklift rental because the
waiting time is only 21 minutes and does not exceed 3
Fig. 10 shows the results of the proposed hours, no rental fee is charged.
improvement, output 1 (a). After the simulation is Cost of waiting (CW) Calculation of service costs for
carried out, it could be seen from the simulation that the the addition of 1 helper and 1 forklift operator Plus
18 trucks that have come could be accommodated with forklift rental
an average waiting time for unloading trucks to storage Total = E (CW) = CS+CW = (Rp. 35,416 x 4 = Rp.
of 0.2205 hours, or about 13.2 minutes, with a utility 141,664) + Rp. 7,000,000
level value of 50%. (b) shows the result of the proposed = Rp. 141,664 +Rp. 7,000,000 = Rp. 7,141,664
improvement 2. It could be seen from the simulation that
there are 18 trucks with an average waiting time for b. Simulation of proposed truck unloading with transit
area providing outdoor racks because the waiting time is
13.2 minutes and does not exceed 3 hours, no rental fee 7. Yuniawan D, Fajar P A, Hariyanto S and
is charged.waiting cost (CW) calculation of service Setiawan R 2018 Traffic queue modeling using
costs for the addition of 1 helper because the capacity arena simulation software (a case study of
for 18 trucks is 2 outdoor shelves so 2 x Rp. 2,000,000 Mergan 4-Way intersection in Malang City) In
=Rp. 4,000,000 MATEC Web of Conferences 204(02009) EDP
Total E (CW) = CS +CW (Rp. 35,416 x 3 = Rp. 106,248) Sciences
+ Rp. 4,000,000) = Rp. 4,106,248 8. Banks J S 2013 Disscrete even system
simulation book pearson education ltd
9. Kelton W D, Sadowski R P and Zupick N B
2015 Simulation With Arena. London US
4 Conclusion McGtaw-HILL europe
The results for the highest data in June, experienced 10. Muravev D, Aksoy S, Rakhmangulov A and
queues due to the length of unloading trucks Aydogdu YV 2016 Comparing model
experiencing overnight charges. The waiting time for development in discrete event simulation on
the truck is up to 4.4 hours using single-channel single- Ro-Ro terminal example. International Journal
phase. With costs reaching Rp. 7,330,832. In the of Logistics Systems and Management
existing model, there is a waiting time of up to 4.4 hours. 24(3)283-297
This affects the work of trucks, which results in 11. Sugiarto F and Buliali J L 2012 Implementasi
trucks not being able to enter or being unable to be simulasi sistem untuk optimasi proses produksi
worked on up to 2 trucks. The proposed improvement pada perusahaan pengalengan ikan. Jurnal
scenario 2 is better than the proposed improvement teknik ITS 1(1) A236-A241
scenario 1 because, as could be seen from the costs 12. Robert G S 2017 Verification and Simulation
incurred by the company in the proposed improvement Model book Winter simulation Conference
scenario 2, with the addition of 1 helper and outdoor 13. Yin C and McKay A 2018. Introduction to
shelves, it only costs Rp. 4,106,248 while the proposed modeling and simulation techniques In
improvement scenario 1 costs Rp. 7,141,664. Proceedings of ISCIIA 2018 and ITCA 2018
Referring to the conclusions of the research that has Leeds
been done, the authors provide suggestions so that the 14. Yuniawan D, Fajar P A, Hariyanto S and
company continues to have good performance and could Setiawan R 2018 Traffic queue modeling using
be more efficient. Each worker is paid more attention arena simulation software (a case study of
and the operator must be swift and responsive if he Mergan 4-Way intersection in Malang City) In
knows the truck is coming. The company PT. XYZ MATEC Web of Conferences (204) 02009 DP
could evaluate and budget for infrastructure needs and Sciences
is expected to conduct a trial of the long waiting time at 15. Wijaya H, Ridwan AY and Setyawan, E B
unloading time. 2022 Designing simulation model for
minimizing coal train unloading time : A case
study of Kereta Api Logistik Company. In IOP
References Conference Series Materials Science and
Engineering (1212)1 012049
1. Sipahutar J A 2020 Analysis of Customer
Queuing System of Bank Rakyat Indonesia
balige branch using muti channel single phase
usu 28445
2. Cipta H 2017 Simulation Of Queue Patient
Service ZERO Jurnal Sains Matematika dan
Terapan 1(1) 22-32
3. Wardana D H and Herlina H 2023 Simulasi
Sistem Pelayanan Bongkar Muat Kapal Untuk
Mengurangi Waiting Time Menggunakan
Sofware Arena Jurnal Taguchi Jurnal Ilmiah
Teknik dan Manajemen Industri 3(1) 53-67
4. Barry R, Hezser J and Chu M 2005 Operations
management Salemba Empat Jakarta
5. Kusuma D D, Wahyudin W and Anshari A
2023 Analisis Teori Antrian dan Optimalisasi
Pelayanan Pada Alfamart Perum Cengkong
Menggunakan Model Single Channel-Single
Phase. Jurnal Serambi Engineering 8(2)
6. Maharani A and Rabbani LS 2020
Identification of risk events and risk agents in
goods hoarding at the port using fishbone
diagram Tibuana, 3(01) 24-29