Search for collections on Undip Repository

PENGEMBANGAN MODEL SISTEM OMNICHANNEL UNTUK PREDIKSI PENJUALAN BERDASARKAN PERILAKU KONSUMEN MENGGUNAKAN METODE PROCESS DISCOVERY ALGORITHM

TRIDALESTARI, Ferra Arik and Mustafid, Mustafid and Jie, Ferry (2024) PENGEMBANGAN MODEL SISTEM OMNICHANNEL UNTUK PREDIKSI PENJUALAN BERDASARKAN PERILAKU KONSUMEN MENGGUNAKAN METODE PROCESS DISCOVERY ALGORITHM. Doctoral thesis, UNIVERSITAS DIPONEGORO.

[thumbnail of COVER dll.pdf] Text
COVER dll.pdf
Restricted to Repository staff only

Download (8MB)
[thumbnail of 9. BAB I - PENDAHULUAN.pdf] Text
9. BAB I - PENDAHULUAN.pdf

Download (726kB)
[thumbnail of 10. BAB II - KAJIAN PUSTAKA & LANDASAN TEORI.pdf] Text
10. BAB II - KAJIAN PUSTAKA & LANDASAN TEORI.pdf

Download (2MB)
[thumbnail of 11. BAB III - METODOLOGI PENELITIAN.pdf] Text
11. BAB III - METODOLOGI PENELITIAN.pdf
Restricted to Repository staff only

Download (1MB)
[thumbnail of 12. BAB IV - HASIL PENELITIAN & PEMBAHASAN.pdf] Text
12. BAB IV - HASIL PENELITIAN & PEMBAHASAN.pdf
Restricted to Repository staff only

Download (4MB)
[thumbnail of 13.BAB V - KESIMPULAN.pdf] Text
13.BAB V - KESIMPULAN.pdf
Restricted to Repository staff only

Download (670kB)
[thumbnail of 14. DAFTAR PUSTAKA.pdf] Text
14. DAFTAR PUSTAKA.pdf

Download (650kB)
[thumbnail of 15. LAMPIRAN.pdf] Text
15. LAMPIRAN.pdf
Restricted to Repository staff only

Download (2MB)

Abstract

Di era Society 5.0, teknologi seperti Internet of Things (IoT), Kecerdasan Buatan (AI), dan Big Data memainkan peran penting dalam kehidupan sehari-hari, termasuk dalam aktivitas e-commerce. Salah satu strategi e-commerce yang berkembang adalah penerapan sistem omnichannel, yang mengintegrasikan saluran penjualan online dan offline untuk memberikan pengalaman berbelanja yang mulus bagi konsumen. Namun, pengelolaan data transaksi yang kompleks dan pemahaman terhadap perilaku konsumen yang dinamis menjadi tantangan bagi bisnis yang menerapkan sistem omnichannel. Penelitian saat ini lebih banyak berfokus pada model konseptual dan implementasi omnichannel, dengan sedikit yang membahas analisis data penjualan. Penelitian ini bertujuan untuk mengembangkan model sistem omnichannel untuk prediksi penjualan berdasarkan data perilaku konsumen. Pendekatan Soft System Methodology (SSM) digunakan dalam penelitian ini untuk merancang sistem omnichannel berbasis prediksi. Metode hybrid yang mengombinasikan Algoritma Process Discovery untuk menganalisis perilaku konsumen dan ARIMAX untuk prediksi penjualan berdasarkan aktivitas konsumen diterapkan. Hasil penelitian menunjukkan bahwa SSM efektif dalam mengembangkan model prediksi pada sistem omnichannel, yang berfokus pada empat proses utama: interaksi konsumen, operasional sistem omnichannel, prediksi penjualan, dan rekomendasi. Model prediksi ini diawali dengan analisis model proses yang dihasilkan oleh Algoritma Fuzzy Miner, yang terbukti efektif dalam membaca perilaku konsumen di berbagai saluran utama, seperti Media Sosial, Marketplace, Toko di Media Sosial, Webstore, dan platform Messenger. Model proses ini memberikan rekomendasi mengenai saluran yang harus diprioritaskan dan saluran mana yang memberikan kontribusi terbesar terhadap penjualan, yang diperoleh dari frekuensi aktivitas deskriptif dan korelasi antar saluran. Sementara itu, analisis dengan pendekatan ARIMAX menghasilkan model prediksi penjualan yang menunjukkan saluran dengan pengaruh terbesar terhadap nilai penjualan berdasarkan aktivitas konsumen setiap hari di berbagai saluran.
Kata kunci: Sistem omnichannel, perilaku konsumen, prediksi penjualan, Process Discovery Algorithm, ARIMAX

In the era of Society 5.0, technologies such as the Internet of Things (IoT), Artificial Intelligence (AI), and Big Data play an important role in daily life, including in e-commerce activities. One of the emerging e-commerce strategies is the implementation of an omnichannel system, which integrates online and offline sales channels to provide a seamless shopping experience for consumers. However, managing complex transaction data and understanding dynamic consumer behavior pose challenges for businesses implementing an omnichannel system. Current research focuses more on conceptual models and omnichannel implementation, with little discussion on sales data analysis. This research aims to develop an omnichannel system model for sales prediction based on consumer behavior data. The Soft System Methodology (SSM) approach is used in this research to design a prediction-based omnichannel system. A hybrid method that combines the Process Discovery Algorithm to analyze consumer behavior and ARIMAX for sales prediction based on consumer activity is applied. The research results show that SSM is effective in developing predictive models in omnichannel systems, focusing on four main processes: consumer interaction, omnichannel system operations, sales prediction, and recommendations. This predictive model begins with process model analysis generated by the Fuzzy Miner Algorithm, which has proven effective in reading consumer behavior across major channels such as Social Media, Marketplaces, Social Media Stores, Webstores, and Messenger platforms. This process model provides recommendations on which channels should be prioritized and which channels contribute the most to sales, based on the frequency of descriptive activities and correlations between channels. Meanwhile, analysis using the ARIMAX approach produces a sales prediction model that shows the channels with the greatest influence on sales value based on daily consumer activity across various channels.
Keywords: Omnichannel system, consumer behavior, sales prediction, Process Discovery Algorithm, ARIMAX

Item Type: Thesis (Doctoral)
Uncontrolled Keywords: Sistem omnichannel, perilaku konsumen, prediksi penjualan, Process Discovery Algorithm, ARIMAX
Subjects: Sciences and Mathemathic
Divisions: Postgraduate Program > Doctor Program in Information System
Depositing User: ekana listianawati
Date Deposited: 21 Apr 2025 06:31
Last Modified: 21 Apr 2025 06:31
URI: https://eprints2.undip.ac.id/id/eprint/31305

Actions (login required)

View Item View Item