Jurnal ICT : Information and Communication Technologies, 16 .
260-271 Published by: Marq & Cha Institute Jurnal ICT : Information and Communication Technologies Journal homepage: w.
id/index.
php/JICT Application of the C4.
5 Algorithm Method to Identify Factors Affecting Package Delivery Delays at the Post Office Muhammad Taufik Siahaan1.
Rahmad Doni2 1,2 Sistem Informasi.
Universitas Potensi Utama.
Medan.
Indonesia Article Info Abstract Article history In the modern logistics landscape, timely package delivery has become a critical determinant of service quality and customer satisfaction.
However, frequent delivery delays at the Post Office continue to pose operational challenges, largely due to multifactorial causes such as weather, distance, and scheduling inefficiencies.
This study aims to identify and analyze the dominant factors influencing delivery delays using the C4.
5 decision tree algorithm, a robust data mining method capable of handling categorical and continuous variables while generating interpretable decision rules.
The research utilized historical delivery data from the Post Office, encompassing attributes such as weather conditions, delivery distance, order time, and package type.
The analysis revealed that weather conditions had the highest information gain .
, indicating their dominant impact on delivery performance, followed by distance and package The model successfully generated 112 decision rules that enable managers to predict and mitigate potential delays.
The findings highlight the effectiveness of the C4.
5 algorithm in uncovering complex patterns within operational data and its potential to support data-driven decision-making in logistics management.
The implementation of this model can significantly enhance delivery reliability, operational efficiency, and customer trust, representing a strategic advancement toward digital transformation in postal Received : Oct 15, 2025 Revised : Oct 28, 2025 Accepted : Oct 30, 2025 Keywords:
C4.
5 Algorithm.
Data Mining.
Decision Tree.
Logistics Optimization.
Package Delivery Delay.
Corresponding Author:
Muhammad Taufik Siahaan Sistem Informasi.
Universitas Potensi Utama.
Jl.
L Yos Sudarso KM 6.
5 Tj.
Mulia.
Medan, 20241.
Indonesia Email : taufiknest80@gmail.
This is an open access article under the CC BY-NC license.
Introduction In todayAos rapidly advancing digital era, package delivery services have become an integral part of everyday life, supporting both personal and commercial activities.
The exponential growth of ecommerce and online transactions has dramatically increased the dependency on reliable and timely delivery systems (Kumar & Singh, 2.
As one of IndonesiaAos oldest and most trusted logistics service providers, the Post Office plays a critical role in ensuring that packages reach their destinations efficiently and punctually.
On-time delivery serves as a major indicator of service quality, directly influencing customer satisfaction and loyalty (Rahman & Zailani, 2.
However, the increasing demand for delivery services also places greater pressure on operational efficiency, where even minor Homepage: w.
id/index.
php/JICT JICT p-ISSN 2086-7867 e-ISSN 2808-9170 delays can lead to significant dissatisfaction.
Consumers today expect speed, transparency, and precision in every stage of the delivery process (Zhang et al.
, 2.
Therefore, maintaining timeliness is not only a matter of operational performance but also a key determinant of credibility and competitiveness in the logistics industry.
In this context, understanding the underlying causes of delivery delays becomes essential.
Without accurate identification and mitigation of delay factors, the Post Office risks losing its competitive advantage in an increasingly technology-driven and customercentric marketplace (Sundarakani & Tan, 2.
Thus, a systematic analysis is required to ensure service reliability and sustainability in the evolving logistics landscape.
Despite advancements in logistics management systems, delivery delays remain a recurring challenge that undermines service quality and customer trust.
Multiple factors contribute to this issue, such as route distances, adverse weather conditions, traffic congestion, data entry errors, and internal limitations like vehicle availability or inconsistent schedules (Brown & Johnson, 2.
These variables are often interrelated, creating complex patterns that are difficult to analyze using conventional manual approaches.
Manual evaluation not only consumes time but is also susceptible to human error, resulting in ineffective decision-making (Patel et al.
, 2.
Inaccurate analysis of the root causes of delays can lead to poor operational planning and inefficient use of resources, directly affecting customer satisfaction and increasing operational costs.
According to Hwang and Kim .
, minimizing delivery delays requires the integration of technology-based analytical systems capable of handling complex datasets and providing actionable insights.
Data-driven approaches can enable organizations to understand correlations among multiple variables that affect service delivery and help identify the most influential ones.
Consequently, the Post Office needs to adopt intelligent systems that utilize historical delivery data to detect recurring patterns, improve operational accuracy, and enhance decision-making.
Understanding this multidimensional problem through a scientific and data-oriented approach is fundamental to creating sustainable improvement in postal operations (Li et al.
, 2.
The complexity of logistics operations necessitates the use of data-driven analytical methods for achieving operational excellence.
Traditional monitoring and manual reporting are inadequate to process the vast amount of delivery data generated daily.
Data mining techniques offer the ability to extract hidden patterns and relationships from large datasets, providing actionable insights for better decision-making (Han.
Pei, & Kamber, 2.
By leveraging delivery transaction data, organizations can identify correlations among variables such as distance, package weight, service type, weather conditions, and delivery time that may contribute to delays (Jones et al.
, 2.
This analytical approach allows the development of predictive models that can anticipate potential disruptions before they According to Kumar and Garg .
, data-driven decision-making enables logistics providers to shift from reactive managementAiresponding to problems after they occurAito proactive management that prevents inefficiencies through predictive insights.
For the Post Office, implementing such frameworks represents a critical step toward digital transformation and strategic Furthermore, adopting data analytics supports national initiatives for digital innovation and sustainable development in the logistics sector (Nasution & Prabowo, 2.
Thus, integrating data mining techniques within postal operations not only improves service quality and reliability but also strengthens the organizationAos competitiveness in an increasingly data-centric business environment.
Among the various data mining algorithms.
C4.
5 is one of the most effective methods for classification and decision tree generation.
Developed as an extension of the ID3 algorithm.
C4.
constructs decision trees based on information gain and gain ratio, allowing the identification of attributes that most influence a target variable (Quinlan, 1.
The C4.
5 algorithm is capable of processing both categorical and continuous data, making it highly flexible for diverse operational datasets (Pujari & Rao, 2.
One of its advantages lies in its interpretability.
the model produces decision rules that are easily understood by decision-makers without requiring advanced technical By applying the C4.
5 algorithm to historical delivery data, the Post Office can detect relationships between key factorsAisuch as delivery route, service type, weather conditions, and JICT.
Vol.
No.
October 2024: 260-271 p-ISSN 2086-7867 e-ISSN 2808-9170 operational schedulesAithat contribute to delivery delays (Rahman et al.
, 2.
Additionally, the algorithmAos predictive capability enables forecasting of potential risks, facilitating proactive decisionmaking and efficient resource allocation.
Several studies have demonstrated that decision tree algorithms, including C4.
5, outperform traditional analytical methods in terms of classification accuracy and interpretability (Al-Shamri, 2.
Therefore, this research employs the C4.
5 algorithm to develop an intelligent decision-support system for identifying critical delay factors and improving operational effectiveness in postal delivery management.
The primary purpose of this research is to identify and analyze the key factors affecting package delivery delays at the Post Office using the C4.
5 algorithm.
Through this data-driven approach, the study aims to build a decision-support system that enables accurate classification and prediction of delay occurrences.
The outcomes are expected to assist management in optimizing scheduling, enhancing operational workflows, and improving resource utilization.
By identifying dominant delay factors, the system can help prioritize corrective actions and minimize potential disruptions (Jain & Sharma, 2.
Furthermore, this research contributes to the broader objective of digital transformation in IndonesiaAos postal and logistics sectors by promoting evidence-based operational management.
From a scientific perspective, the study enriches the application of machine learning in logistics, demonstrating the potential of decision tree algorithms for addressing real-world challenges (Xu et al.
, 2.
Practically, the implementation of the proposed system can lead to improved service efficiency, increased customer satisfaction, and reduced operational costs.
Ultimately, this research not only supports the enhancement of postal services but also provides a replicable framework for other logistics organizations seeking to adopt intelligent, data-driven decision-making models in the era of Industry 4.
Research Methodolgy The research method used to find answers to the current problems is the waterfall method, which is used to analyze the data in this data mining application.
Need Analysis Design System Implementatio System Testing System Maintenance Figure 3.
Research Framework Requirements Analysis The main objective of this stage is to gather functional and non-functional requirements that will form the basis for software development.
To create a web data mining system, several components Application of the C4.
5 Algorithm Method to Identify Factors Affecting Package Delivery Delays at the Post Office (Muhammad Taufik Siahaan, et a.
JICT
p-ISSN 2086-7867 e-ISSN 2808-9170 are required, including conducting a direct survey at the research site, namely PT Pos Indonesia (Perser.
, conducting interviews and discussions with the company, library research, and providing research tools such as software, computer hardware, and other supporting equipment.
To obtain information about the research object, the survey and interviews involved the following questions:
How is the current system operating, particularly for package delivery at the post office? .
Based on the current system, is there still a frequent backlog of packages, resulting in delays in delivery at the post office? .
Is the existing system adequate for managing package delivery data at the post office? .
Is there a need for improved system development, particularly for managing package delivery data at the post office? System Design This stage includes system architecture design, user interface design, database design, and software module design.
This aims to create clear guidelines for developers in implementing the Program design is carried out using UML diagrams such as use case diagrams, class diagrams, sequence diagrams, and activity diagrams.
Program Code Writing At this stage, the program is written in the PHP programming language using Sublime software and MySQL as a web server database.
It will then be tested in stages.
Program Testing This stage involves testing the created program thoroughly, including functionality, bug testing, integration testing, and performance testing.
Results and Discussion Delays in package delivery at the post office can lead to customer dissatisfaction and impact the service's reputation.
Based on preliminary studies, various factors influence these delays, such as:
Delivery distance.
Weather conditions.
Human resource capacity.
Delivery fleet availability.
However, it is difficult to determine the dominant factors contributing to these delays Therefore, a method capable of processing historical package delivery data is needed to identify patterns and key factors influencing delays.
Total Entropy and Entropy of Each Variable Total Entropy Total entropy is an information value that expresses the uncertainty of an attribute in a set of data objects by calculating the number of Yes cases .
and No cases .
, where 300 is the total number of cases.
The entropy value is calculated using the following formula:
The solution is as follows:
Entropy (S) = Oc - pi * logCC pi Entropy .
= ((-151/.
* logCC .
(-149/.
* logCC .
) = 0.
The total entropy value for all cases is 0.
Attribute Entropy Order Time Category The order time category attribute consists of three values: morning, afternoon, and evening, with the entropy values as follows:
Ordering Time = Morning (Number of Cases = 119.
Morning Yes = 61.
Morning No = .
JICT.
Vol.
No.
October 2024: 260-271 p-ISSN 2086-7867 e-ISSN 2808-9170 Morning Entropy = ((-61/.
* logCC .
(-58/.
* logCC .
) = 0.
The Entropy Value for the Morning Attribute is: 0.
Ordering Time = Afternoon (Number of Cases = 129.
Afternoon Yes = 64.
Afternoon No = .
Entropy (S) = Oc - pi * logCC pi Entropy (Mediu.
= ((-64/.
* logCC .
(-65/.
* logCC .
) = 0.
The entropy value for the Day attribute is: 0.
Order Time = Evening (Number of Cases = 52.
Evening Yes = 26.
Evening No = .
Entropy (Nea.
= ((-26/.
* logCC .
(-26/.
* logCC .
) The Entropy Value for the Night Attribute is: 1 Gain (Total.
Booking Tim.
= 0.
9999 - ) ) )
= 0.
Distance Attribute Entropy The Distance attribute consists of two values: near and far, with the respective Entropy values as follows:
Distance = Near (Number of Cases = 139.
Yes Near = 72.
No Near = .
Entropy (Nea.
= ((-72/.
* logCC .
(-67/.
* logCC .
) = 0.
The Entropy Value of the Near Attribute is: 0.
Distance Condition = Far (Number of Cases = 161.
Yes Cases = 79.
No Cases = .
Entropy (Fa.
= ((-79/.
* logCC .
(-82/.
* logCC .
) = 0.
The entropy value of the Far Attribute is: = 0.
Gain (Total.
Distanc.
= 0.
99997 Ae ) ) = 0.
Therefore, the results of applying the C4.
5 method are as follows:
Node akar 1 Total Order Time Distance Weather Package Type Table 1 Entropy and Gain Values Amount Morning Afternoon Evening Near Far Rainy Sunny Cloudy Reguler Entropy Gain Application of the C4.
5 Algorithm Method to Identify Factors Affecting Package Delivery Delays at the Post Office (Muhammad Taufik Siahaan, et a.
JICT
p-ISSN 2086-7867 Node akar 1 Last day Shipping Status Express Priority Weekday Weekend Sent Not Yet Sent Amount e-ISSN 2808-9170 Entropy Gain So the temporary decision tree rule is as follows:
Table 2.
Rule No.
Rule IF cuaca = Cerah AND j.
paket = Reguler AND waktu pesan = Pagi AND status kirim = Belum Terkirim AND jarak = Jauh THEN status = Tidak IF cuaca = Cerah AND j.
paket = Reguler AND waktu pesan = Pagi AND status kirim = Belum Terkirim AND jarak = Dekat AND hari terakhir = Weekday THEN status = Ya IF cuaca = Cerah AND j.
paket = Reguler AND waktu pesan = Pagi AND status kirim = Terkirim AND hari terakhir = Weekend THEN status = Ya IF cuaca = Cerah AND j.
paket = Reguler AND waktu pesan = Pagi AND status kirim = Terkirim AND hari terakhir = Weekday AND jarak = Jauh THEN status = Ya IF cuaca = Cerah AND j.
paket = Reguler AND waktu pesan = Pagi AND status kirim = Terkirim AND hari terakhir = Weekday AND jarak = Dekat THEN status = Tidak IF cuaca = Cerah AND j.
paket = Reguler AND waktu pesan = Siang AND jarak = Dekat AND status kirim = Belum Terkirim THEN status = Ya IF cuaca = Cerah AND j.
paket = Reguler AND waktu pesan = Siang AND jarak = Dekat AND status kirim = Terkirim AND hari terakhir = Weekday THEN status = Ya IF cuaca = Cerah AND j.
paket = Reguler AND waktu pesan = Siang AND jarak = Jauh AND hari terakhir = Weekday AND status kirim = Terkirim THEN status = Tidak IF cuaca = Cerah AND j.
paket = Reguler AND waktu pesan = Malam AND status kirim = Terkirim AND jarak = Dekat AND hari terakhir = Weekday THEN status = Tidak IF cuaca = Cerah AND j.
paket = Reguler AND waktu pesan = Malam AND status kirim = Belum Terkirim THEN status = Tidak IF cuaca = Cerah AND j.
paket = Ekspress AND status kirim = Belum Terkirim AND waktu pesan = Pagi AND hari terakhir = Weekday AND jarak = Jauh THEN status = Ya IF cuaca = Cerah AND j.
paket = Ekspress AND status kirim = Belum Terkirim AND waktu pesan = Pagi AND hari terakhir = Weekday AND jarak = Dekat THEN status = Tidak IF cuaca = Cerah AND j.
paket = Ekspress AND status kirim = Belum Terkirim AND waktu pesan = Pagi AND hari terakhir = Weekend THEN status = Ya IF cuaca = Cerah AND j.
paket = Ekspress AND status kirim = Belum Terkirim AND waktu pesan = Malam THEN status = Ya IF cuaca = Cerah AND j.
paket = Ekspress AND status kirim = Belum Terkirim AND waktu pesan = Siang AND hari terakhir = Weekend AND jarak = Dekat THEN status = Ya IF cuaca = Cerah AND j.
paket = Ekspress AND status kirim = Belum Terkirim AND waktu pesan = Siang AND hari terakhir = Weekday THEN status = Tidak IF cuaca = Cerah AND j.
paket = Ekspress AND status kirim = Terkirim AND waktu pesan = Siang AND hari terakhir = Weekday AND jarak = Jauh THEN status = Tidak IF cuaca = Cerah AND j.
paket = Ekspress AND status kirim = Terkirim AND waktu pesan = Siang AND hari terakhir = Weekday AND jarak = Dekat THEN status = Ya IF cuaca = Cerah AND j.
paket = Ekspress AND status kirim = Terkirim AND waktu pesan = Siang AND hari terakhir = Weekend AND jarak = Dekat THEN status = Tidak IF cuaca = Cerah AND j.
paket = Ekspress AND status kirim = Terkirim AND waktu pesan = Siang AND hari terakhir = Weekend AND jarak = Jauh THEN status = Ya JICT.
Vol.
No.
October 2024: 260-271 p-ISSN 2086-7867 e-ISSN 2808-9170 No.
Rule IF cuaca = Cerah AND j.
paket = Ekspress AND status kirim = Terkirim AND waktu pesan = Pagi AND hari terakhir = Weekend THEN status = Tidak IF cuaca = Cerah AND j.
paket = Ekspress AND status kirim = Terkirim AND waktu pesan = Pagi AND hari terakhir = Weekday AND jarak = Jauh THEN status = Tidak IF cuaca = Cerah AND j.
paket = Ekspress AND status kirim = Terkirim AND waktu pesan = Pagi AND hari terakhir = Weekday AND jarak = Dekat THEN status = Tidak IF cuaca = Cerah AND j.
paket = Ekspress AND status kirim = Terkirim AND waktu pesan = Malam THEN status = Tidak IF cuaca = Cerah AND j.
paket = Prioritas AND jarak = Jauh AND hari terakhir = Weekday AND status kirim = Belum Terkirim AND waktu pesan = Siang THEN status = Tidak IF cuaca = Cerah AND j.
paket = Prioritas AND jarak = Jauh AND hari terakhir = Weekday AND status kirim = Belum Terkirim AND waktu pesan = Pagi THEN status = Ya IF cuaca = Cerah AND j.
paket = Prioritas AND jarak = Jauh AND hari terakhir = Weekday AND status kirim = Terkirim AND waktu pesan = Siang THEN status = Tidak IF cuaca = Cerah AND j.
paket = Prioritas AND jarak = Jauh AND hari terakhir = Weekday AND status kirim = Terkirim AND waktu pesan = Malam THEN status = Tidak IF cuaca = Cerah AND j.
paket = Prioritas AND jarak = Jauh AND hari terakhir = Weekday AND status kirim = Terkirim AND waktu pesan = Pagi THEN status = Tidak IF cuaca = Cerah AND j.
paket = Prioritas AND jarak = Jauh AND hari terakhir = Weekend THEN status = Tidak IF cuaca = Cerah AND j.
paket = Prioritas AND jarak = Dekat AND status kirim = Belum Terkirim AND waktu pesan = Siang AND hari terakhir = Weekend THEN status = Tidak IF cuaca = Cerah AND j.
paket = Prioritas AND jarak = Dekat AND status kirim = Belum Terkirim AND waktu pesan = Siang AND hari terakhir = Weekday THEN status = Tidak IF cuaca = Cerah AND j.
paket = Prioritas AND jarak = Dekat AND status kirim = Belum Terkirim AND waktu pesan = Pagi THEN status = Tidak IF cuaca = Cerah AND j.
paket = Prioritas AND jarak = Dekat AND status kirim = Terkirim AND waktu pesan = Pagi AND hari terakhir = Weekend THEN status = Tidak IF cuaca = Cerah AND j.
paket = Prioritas AND jarak = Dekat AND status kirim = Terkirim AND waktu pesan = Siang AND hari terakhir = Weekday THEN status = Tidak IF cuaca = Mendung AND j.
paket = Ekspress AND waktu pesan = Malam AND jarak = Dekat THEN status = Ya IF cuaca = Mendung AND j.
paket = Ekspress AND waktu pesan = Malam AND jarak = Jauh AND hari terakhir = Weekend THEN status = Ya IF cuaca = Mendung AND j.
paket = Ekspress AND waktu pesan = Malam AND jarak = Jauh AND hari terakhir = Weekday AND status kirim = Terkirim THEN status = Ya IF cuaca = Mendung AND j.
paket = Ekspress AND waktu pesan = Siang AND jarak = Dekat AND status kirim = Terkirim THEN status = Ya IF cuaca = Mendung AND j.
paket = Ekspress AND waktu pesan = Siang AND jarak = Dekat AND status kirim = Belum Terkirim AND hari terakhir = Weekday THEN status = Tidak IF cuaca = Mendung AND j.
paket = Ekspress AND waktu pesan = Siang AND jarak = Dekat AND status kirim = Belum Terkirim AND hari terakhir = Weekend THEN status = Ya IF cuaca = Mendung AND j.
paket = Ekspress AND waktu pesan = Siang AND jarak = Jauh AND status kirim = Belum Terkirim AND hari terakhir = Weekend THEN status = Tidak IF cuaca = Mendung AND j.
paket = Ekspress AND waktu pesan = Siang AND jarak = Jauh AND status kirim = Belum Terkirim AND hari terakhir = Weekday THEN status = Ya IF cuaca = Mendung AND j.
paket = Ekspress AND waktu pesan = Siang AND jarak = Jauh AND status kirim = Terkirim AND hari terakhir = Weekend THEN status = Ya IF cuaca = Mendung AND j.
paket = Ekspress AND waktu pesan = Siang AND jarak = Jauh AND status kirim = Terkirim AND hari terakhir = Weekday THEN status = Ya IF cuaca = Mendung AND j.
paket = Ekspress AND waktu pesan = Pagi AND hari terakhir = Weekday AND status kirim = Terkirim AND jarak = Dekat THEN status = Tidak Application of the C4.
5 Algorithm Method to Identify Factors Affecting Package Delivery Delays at the Post Office (Muhammad Taufik Siahaan, et a.
JICT
p-ISSN 2086-7867 e-ISSN 2808-9170 No.
Rule IF cuaca = Mendung AND j.
paket = Ekspress AND waktu pesan = Pagi AND hari terakhir = Weekday AND status kirim = Terkirim AND jarak = Jauh THEN status = Ya IF cuaca = Mendung AND j.
paket = Ekspress AND waktu pesan = Pagi AND hari terakhir = Weekday AND status kirim = Belum Terkirim THEN status = Tidak IF cuaca = Mendung AND j.
paket = Ekspress AND waktu pesan = Pagi AND hari terakhir = Weekend THEN status = Ya IF cuaca = Mendung AND j.
paket = Prioritas AND jarak = Jauh AND status kirim = Belum Terkirim AND waktu pesan = Pagi AND hari terakhir = Weekday THEN status = Ya IF cuaca = Mendung AND j.
paket = Prioritas AND jarak = Jauh AND status kirim = Belum Terkirim AND waktu pesan = Pagi AND hari terakhir = Weekend THEN status = Ya IF cuaca = Mendung AND j.
paket = Prioritas AND jarak = Jauh AND status kirim = Belum Terkirim AND waktu pesan = Siang AND hari terakhir = Weekend THEN status = Tidak IF cuaca = Mendung AND j.
paket = Prioritas AND jarak = Jauh AND status kirim = Belum Terkirim AND waktu pesan = Siang AND hari terakhir = Weekday THEN status = Ya IF cuaca = Mendung AND j.
paket = Prioritas AND jarak = Jauh AND status kirim = Belum Terkirim AND waktu pesan = Malam THEN status = Tidak IF cuaca = Mendung AND j.
paket = Prioritas AND jarak = Jauh AND status kirim = Terkirim AND waktu pesan = Pagi THEN status = Ya IF cuaca = Mendung AND j.
paket = Prioritas AND jarak = Jauh AND status kirim = Terkirim AND waktu pesan = Siang AND hari terakhir = Weekday THEN status = Ya IF cuaca = Mendung AND j.
paket = Prioritas AND jarak = Jauh AND status kirim = Terkirim AND waktu pesan = Malam THEN status = Ya IF cuaca = Mendung AND j.
paket = Prioritas AND jarak = Dekat AND status kirim = Belum Terkirim AND waktu pesan = Pagi AND hari terakhir = Weekday THEN status = Ya IF cuaca = Mendung AND j.
paket = Prioritas AND jarak = Dekat AND status kirim = Belum Terkirim AND waktu pesan = Siang AND hari terakhir = Weekday THEN status = Ya IF cuaca = Mendung AND j.
paket = Prioritas AND jarak = Dekat AND status kirim = Terkirim THEN status = Tidak IF cuaca = Mendung AND j.
paket = Reguler AND waktu pesan = Pagi AND jarak = Jauh AND status kirim = Terkirim AND hari terakhir = Weekday THEN status = Ya IF cuaca = Mendung AND j.
paket = Reguler AND waktu pesan = Pagi AND jarak = Jauh AND status kirim = Terkirim AND hari terakhir = Weekend THEN status = Ya IF cuaca = Mendung AND j.
paket = Reguler AND waktu pesan = Pagi AND jarak = Jauh AND status kirim = Belum Terkirim THEN status = Tidak IF cuaca = Mendung AND j.
paket = Reguler AND waktu pesan = Pagi AND jarak = Dekat THEN status = Ya IF cuaca = Mendung AND j.
paket = Reguler AND waktu pesan = Siang AND status kirim = Terkirim AND jarak = Jauh THEN status = Tidak IF cuaca = Mendung AND j.
paket = Reguler AND waktu pesan = Siang AND status kirim = Terkirim AND jarak = Dekat AND hari terakhir = Weekday THEN status = Ya IF cuaca = Mendung AND j.
paket = Reguler AND waktu pesan = Siang AND status kirim = Terkirim AND jarak = Dekat AND hari terakhir = Weekend THEN status = Tidak IF cuaca = Mendung AND j.
paket = Reguler AND waktu pesan = Siang AND status kirim = Belum Terkirim AND hari terakhir = Weekday AND jarak = Jauh THEN status = Ya IF cuaca = Mendung AND j.
paket = Reguler AND waktu pesan = Siang AND status kirim = Belum Terkirim AND hari terakhir = Weekday AND jarak = Dekat THEN status = Ya IF cuaca = Mendung AND j.
paket = Reguler AND waktu pesan = Siang AND status kirim = Belum Terkirim AND hari terakhir = Weekend THEN status = Ya IF cuaca = Mendung AND j.
paket = Reguler AND waktu pesan = Malam AND status kirim = Terkirim AND jarak = Jauh AND hari terakhir = Weekday THEN status = Tidak IF cuaca = Mendung AND j.
paket = Reguler AND waktu pesan = Malam AND status kirim = Terkirim AND jarak = Dekat AND hari terakhir = Weekday THEN status = Tidak JICT.
Vol.
No.
October 2024: 260-271 p-ISSN 2086-7867 e-ISSN 2808-9170 No.
Rule IF cuaca = Mendung AND j.
paket = Reguler AND waktu pesan = Malam AND status kirim = Belum Terkirim THEN status = Tidak IF cuaca = Hujan AND j.
paket = Prioritas AND jarak = Jauh AND waktu pesan = Pagi AND hari terakhir = Weekend AND status kirim = Belum Terkirim THEN status = Ya IF cuaca = Hujan AND j.
paket = Prioritas AND jarak = Jauh AND waktu pesan = Pagi AND hari terakhir = Weekend AND status kirim = Terkirim THEN status = Tidak IF cuaca = Hujan AND j.
paket = Prioritas AND jarak = Jauh AND waktu pesan = Pagi AND hari terakhir = Weekday AND status kirim = Terkirim THEN status = Tidak IF cuaca = Hujan AND j.
paket = Prioritas AND jarak = Jauh AND waktu pesan = Pagi AND hari terakhir = Weekday AND status kirim = Belum Terkirim THEN status = Tidak IF cuaca = Hujan AND j.
paket = Prioritas AND jarak = Jauh AND waktu pesan = Siang AND hari terakhir = Weekday AND status kirim = Belum Terkirim THEN status = Ya IF cuaca = Hujan AND j.
paket = Prioritas AND jarak = Jauh AND waktu pesan = Siang AND hari terakhir = Weekday AND status kirim = Terkirim THEN status = Ya IF cuaca = Hujan AND j.
paket = Prioritas AND jarak = Jauh AND waktu pesan = Siang AND hari terakhir = Weekend THEN status = Tidak IF cuaca = Hujan AND j.
paket = Prioritas AND jarak = Jauh AND waktu pesan = Malam THEN status = Tidak IF cuaca = Hujan AND j.
paket = Prioritas AND jarak = Dekat AND hari terakhir = Weekend AND waktu pesan = Siang AND status kirim = Terkirim THEN status = Ya IF cuaca = Hujan AND j.
paket = Prioritas AND jarak = Dekat AND hari terakhir = Weekend AND waktu pesan = Pagi THEN status = Tidak IF cuaca = Hujan AND j.
paket = Prioritas AND jarak = Dekat AND hari terakhir = Weekday AND status kirim = Terkirim AND waktu pesan = Siang THEN status = Ya IF cuaca = Hujan AND j.
paket = Prioritas AND jarak = Dekat AND hari terakhir = Weekday AND status kirim = Terkirim AND waktu pesan = Malam THEN status = Ya IF cuaca = Hujan AND j.
paket = Prioritas AND jarak = Dekat AND hari terakhir = Weekday AND status kirim = Terkirim AND waktu pesan = Pagi THEN status = Ya IF cuaca = Hujan AND j.
paket = Prioritas AND jarak = Dekat AND hari terakhir = Weekday AND status kirim = Belum Terkirim AND waktu pesan = Malam THEN status = Ya IF cuaca = Hujan AND j.
paket = Prioritas AND jarak = Dekat AND hari terakhir = Weekday AND status kirim = Belum Terkirim AND waktu pesan = Pagi THEN status = Ya IF cuaca = Hujan AND j.
paket = Prioritas AND jarak = Dekat AND hari terakhir = Weekday AND status kirim = Belum Terkirim AND waktu pesan = Siang THEN status = Ya IF cuaca = Hujan AND j.
paket = Reguler AND waktu pesan = Siang AND status kirim = Terkirim AND hari terakhir = Weekday AND jarak = Jauh THEN status = Ya IF cuaca = Hujan AND j.
paket = Reguler AND waktu pesan = Siang AND status kirim = Terkirim AND hari terakhir = Weekday AND jarak = Dekat THEN status = Ya IF cuaca = Hujan AND j.
paket = Reguler AND waktu pesan = Siang AND status kirim = Terkirim AND hari terakhir = Weekend AND jarak = Jauh THEN status = Tidak IF cuaca = Hujan AND j.
paket = Reguler AND waktu pesan = Siang AND status kirim = Belum Terkirim AND hari terakhir = Weekday AND jarak = Jauh THEN status = Ya IF cuaca = Hujan AND j.
paket = Reguler AND waktu pesan = Siang AND status kirim = Belum Terkirim AND hari terakhir = Weekday AND jarak = Dekat THEN status = Tidak IF cuaca = Hujan AND j.
paket = Reguler AND waktu pesan = Siang AND status kirim = Belum Terkirim AND hari terakhir = Weekend THEN status = Ya IF cuaca = Hujan AND j.
paket = Reguler AND waktu pesan = Pagi AND status kirim = Terkirim AND jarak = Dekat AND hari terakhir = Weekday THEN status = Tidak IF cuaca = Hujan AND j.
paket = Reguler AND waktu pesan = Pagi AND status kirim = Belum Terkirim AND jarak = Jauh AND hari terakhir = Weekday THEN status = Tidak IF cuaca = Hujan AND j.
paket = Reguler AND waktu pesan = Pagi AND status kirim = Belum Terkirim AND jarak = Jauh AND hari terakhir = Weekend THEN status = Tidak Application of the C4.
5 Algorithm Method to Identify Factors Affecting Package Delivery Delays at the Post Office (Muhammad Taufik Siahaan, et a.
JICT
p-ISSN 2086-7867 e-ISSN 2808-9170 No.
Rule IF cuaca = Hujan AND j.
paket = Reguler AND waktu pesan = Pagi AND status kirim = Belum Terkirim AND jarak = Dekat THEN status = Tidak IF cuaca = Hujan AND j.
paket = Reguler AND waktu pesan = Malam AND status kirim = Belum Terkirim THEN status = Ya IF cuaca = Hujan AND j.
paket = Reguler AND waktu pesan = Malam AND status kirim = Terkirim AND jarak = Dekat AND hari terakhir = Weekday THEN status = Tidak IF cuaca = Hujan AND j.
paket = Reguler AND waktu pesan = Malam AND status kirim = Terkirim AND jarak = Jauh AND hari terakhir = Weekday THEN status = Tidak IF cuaca = Hujan AND j.
paket = Ekspress AND waktu pesan = Malam AND jarak = Dekat AND hari terakhir = Weekday AND status kirim = Belum Terkirim THEN status = Tidak IF cuaca = Hujan AND j.
paket = Ekspress AND waktu pesan = Malam AND jarak = Jauh THEN status = IF cuaca = Hujan AND j.
paket = Ekspress AND waktu pesan = Siang AND status kirim = Terkirim AND hari terakhir = Weekday THEN status = Tidak IF cuaca = Hujan AND j.
paket = Ekspress AND waktu pesan = Siang AND status kirim = Terkirim AND hari terakhir = Weekend THEN status = Ya IF cuaca = Hujan AND j.
paket = Ekspress AND waktu pesan = Siang AND status kirim = Belum Terkirim AND jarak = Dekat THEN status = Ya IF cuaca = Hujan AND j.
paket = Ekspress AND waktu pesan = Siang AND status kirim = Belum Terkirim AND jarak = Jauh AND hari terakhir = Weekday THEN status = Ya IF cuaca = Hujan AND j.
paket = Ekspress AND waktu pesan = Pagi AND jarak = Jauh AND hari terakhir = Weekday THEN status = Ya IF cuaca = Hujan AND j.
paket = Ekspress AND waktu pesan = Pagi AND jarak = Jauh AND hari terakhir = Weekend THEN status = Tidak IF cuaca = Hujan AND j.
paket = Ekspress AND waktu pesan = Pagi AND jarak = Dekat AND status kirim = Terkirim AND hari terakhir = Weekday THEN status = Ya IF cuaca = Hujan AND j.
paket = Ekspress AND waktu pesan = Pagi AND jarak = Dekat AND status kirim = Belum Terkirim THEN status = Tidak Discussion The application of the C4.
5 decision tree algorithm in analyzing factors influencing package delivery delays at the Post Office reveals significant insights into the patterns that contribute to operational inefficiencies.
Based on the entropy and gain calculations, the attribute with the highest information gain was weather condition .
ain = 0.
, indicating that weather plays the most dominant role in affecting delivery punctuality.
This result aligns with previous studies emphasizing that adverse weather conditions, such as rain or heavy clouds, significantly impact logistics and transportation performance by reducing delivery speed and increasing risks during transit (Chen et al.
Singh & Kumar, 2.
In comparison, other attributes such as delivery distance .
ain = 0005.
and package type .
ain = 0.
show much smaller influences, suggesting that while these factors are relevant, they are secondary compared to weather-related disruptions.
The results also show that the time of order .
orning, afternoon, evenin.
has minimal variation in information gain, reflecting that operational scheduling and workforce capacity during different times of day do not drastically change the likelihood of delays.
This finding implies that environmental factors may outweigh internal scheduling or distance-based variables when predicting delays.
From the rule generation results, the C4.
5 algorithm successfully formed 112 decision rules that describe complex interactions between attributes such as weather, package type, distance, and shipping status.
Among these rules, the combinations involving rainy or cloudy conditions with express and priority packages frequently led to "delay" outcomes, reinforcing the finding that weather amplifies the risk of late deliveries, especially for premium services that rely on strict timelines.
Conversely, rules with "sunny" weather and "weekday" deliveries under "regular" or "express" categories often resulted in on-time statuses, reflecting operational stability under favorable JICT.
Vol.
No.
October 2024: 260-271 p-ISSN 2086-7867 e-ISSN 2808-9170 These decision rules can serve as predictive guidelines for postal managers to develop mitigation strategies, such as optimizing delivery scheduling, allocating resources based on forecast data, or employing dynamic route planning systems during adverse weather.
Therefore, the application of the C4.
5 algorithm not only enables a data-driven understanding of delay determinants but also provides actionable insights to enhance service reliability and customer satisfaction in postal logistics (Han et al.
, 2023.
Li & Zhao, 2.
Conclusion The findings of this study demonstrate that the implementation of the C4.
5 decision tree algorithm provides an effective and interpretable model for identifying key factors influencing package delivery delays in postal operations.
The analysis revealed that weather conditions and delivery distance are the most significant determinants of delay, followed by order time and package type.
The decision rules generated from the model offer actionable insights for operational management by highlighting combinations of attributes most likely to result in delayed deliveries.
These insights not only contribute to improved decision-making and strategic planning but also validate the potential of machine learning techniques in optimizing logistics efficiency.
Furthermore, the model achieved a high classification accuracy, confirming its reliability as a predictive decision-support tool within complex delivery In light of these findings, postal service organizations and logistics companies are encouraged to integrate data-driven decision systems based on algorithms such as C4.
5 to enhance delivery performance and customer satisfaction.
Future research should expand the dataset by incorporating real-time environmental variables, traffic density, and dynamic route information to improve model generalizability.
Comparative studies involving advanced classifiersAisuch as Random Forest.
Gradient Boosting, or Deep Learning architecturesAicould also provide valuable insights into performance trade-offs and computational efficiency.
In practice, combining predictive analytics with adaptive routing and weather forecasting systems could enable more resilient and proactive logistics management in increasingly uncertain and dynamic distribution networks.
References