Marada, Livia Fragrancy (2026) Pemodelan Jumlah Keluhan Pelanggan terhadap Layanan Sebuah Perusahaan Terminal Peti Kemas di Surabaya menggunakan Regresi Poisson Bayesian dan Regresi Mixture Poisson Bayesian. Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 5003221107-Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
5003221107-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (2MB) |
Abstract
Terminal peti kemas memainkan peran penting dalam rantai pasok global dan, sebagai penyedia layanan B2B, dituntut untuk menjaga hubungan jangka panjang dengan pelanggannya. Pemanfaatan data keluhan pelanggan dalam Customer Relationship Management (CRM) telah menjadi strategi yang penting. Oleh karena itu, penelitian ini menggunakan regresi Poisson Bayesian dan regresi mixture Poisson Bayesian untuk menganalisis data jumlah keluhan pelanggan. Model regresi Bayesian dibangun menggunakan pseudo-prior yang diturunkan dari Generalized Linear Models (GLMs), sedangkan model mixture dibentuk oleh dua komponen yang dibuat berdasarkan klasifikasi pelanggan dalam aktivitas transaksi pelanggan. Model-model tersebut kemudian dievaluasi menggunakan Watanabe-Akaike Information Criterion (WAIC). Hasil penelitian menunjukkan bahwa regresi mixture Poisson Bayesian memberikan kinerja terbaik dengan nilai WAIC paling rendah senilai 2.106,7. Nilai WAIC yang relatif besar dipengaruhi oleh jumlah observasi yang tinggi pada data serta skala log-likelihood yang digunakan dalam estimasi model. Meskipun memiliki struktur yang lebih kompleks, model regresi mixture Poisson mampu memberikan keseimbangan terbaik antara kompleksitas model dan akurasi prediksi dalam memodelkan data count berukuran besar dengan kovariat yang beragam. Model mixture terpilih menunjukkan perlakuan yang berbeda untuk setiap komponen. Hasil ini menyoroti perbedaan karakteristik layanan dan ekspektasi di berbagai bidang bisnis serta mendukung penggunaan model mixture Bayesian dalam menganalisis keluhan pelanggan B2B.
=====================================================================================================================================
Container terminals play an important role in the global supply chain and, as B2B service providers, are required to maintain long-term relationships with the customers. The use of customer complaint data in Customer Relationship Management (CRM) has become an important strategy. Therefore, this study uses Bayesian Poisson Regression and Bayesian Mixture Poisson Regression to analyze customer complaint data. The Bayesian regression model was constructed using pseudo-priors derived from Generalized Linear Models (GLMs), while the mixture model was formed by two components based on customer classification in customer transaction activites. The models were then evaluated using the Watanabe-Akaike Information Criterion (WAIC). The results show that Bayesian Poisson mixture regression provides the best performance with the lowest WAIC value of 2106.7. The relatively large WAIC value is influenced by the high number of observations in the data and the log-likelihood scale used in model estimation. Despite its more complex structure, the Poisson mixture regression model is able to provide the best balance between model complexity and prediction accuracy in modeling large count data with diverse covariates. The selected mixture model shows different treatments for each component. These results highlight the differences in service characteristics and expectations in various business fields and support the use of Bayesian mixture models in analyzing B2B customer complaints.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | Inferensi Bayesian, Regresi Poisson, Regresi Mixture Poisson, Keluhan Pelanggan, Terminal Peti Kemas, Bayesian Inference, Poisson Regression, Poisson Mixture Regression, Customer Complaint, Container Terminal |
| Subjects: | H Social Sciences > HA Statistics H Social Sciences > HA Statistics > HA31.3 Regression. Correlation. Logistic regression analysis. H Social Sciences > HA Statistics > HA31.7 Estimation Q Science > QA Mathematics > QA279.5 Bayesian statistical decision theory. |
| Divisions: | Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis |
| Depositing User: | Livia Fragrancy Marada |
| Date Deposited: | 29 Jan 2026 03:33 |
| Last Modified: | 29 Jan 2026 03:33 |
| URI: | http://repository.its.ac.id/id/eprint/131112 |
Actions (login required)
 |
View Item |