Dampak Lean Terhadap Transformasi Digital: Analisis Maturitas Industri 4.0 Memanfaatkan Pembelajaran Mesin
Keywords:
lean, maturitas industri 4.0, pembelajaran mesin, pohon keputusan, transformasi digital, decision tree, digital transformation, industry 4.0 maturity, machine learningAbstract
Revolusi industri keempat membawa peluang sekaligus tantangan digital baru bagi dunia bisnis. Penelitian ini dilakukan untuk menganalisis status maturitas global Industri 4.0 guna mengidentifikasi tren spesifik, tantangan utama, dan potensi pertumbuhan industri. Teknik pembelajaran mesin tingkat lanjut diterapkan untuk menganalisis data, memproyeksikan peta jalan masa depan, serta memberikan rekomendasi yang disesuaikan demi meningkatkan kemampuan pengambilan keputusan strategis perusahaan. Tahap awal menggunakan Hierarchical Clustering untuk mengelompokkan data secara terstruktur, dilanjutkan dengan evaluasi model Decision Tree (DT), Support Vector Machine, dan Random Forest. Model Decision Tree (DT) terpilih sebagai model dengan performa terbaik dengan nilai MSE 0,032, MAE 0,063, dan skor R2 sebesar 0,862. Hasil analisis kuantitatif menunjukkan bahwa strategi optimasi jalur keputusan mampu meningkatkan level maturitas digital dari 1,177 menjadi 1,382, serta melalui jalur alternatif Efisiensi Waktu Perbaikan Deteksi Pemborosan meningkat signifikan dari 0,754 menjadi 1,42. Terkait pengaruh Lean, ditemukan bahwa variabel lean bukan merupakan faktor dominan yang memengaruhi skor maturitas secara langsung, melainkan lebih berperan dalam menentukan arah strategi implementasi perusahaan. Kontribusi ilmiah utama penelitian ini terletak pada integrasi model interdisipliner Digital Lean melalui algoritma Decision Tree (DT) untuk memproyeksikan estimasi peningkatan skor maturitas digital secara transparan melalui jalur intervensi yang terarah.
The fourth industrial revolution presents distinct digital opportunities and challenges for the business sector. This research assesses the global maturity status of Industry 4.0 to identify industry-specific trends, critical challenges, and growth potential. Advanced machine learning techniques are deployed for data analysis, future roadmap projection, and the provision of tailored recommendations to enhance strategic corporate decision-making. The initial phase leverages Hierarchical Clustering for structured data grouping, followed by the evaluation of Decision Tree (DT), Support Vector Machine, and Random Forest models. The Decision Tree (DT) model was identified as the optimal model, achieving an MSE of 0.032, MAE of 0.063, and an R2 score of 0.862. Quantitative results demonstrate that the proposed path optimization strategy can elevate the digital maturity level from 1.177 to 1.382, while an alternative path focusing on Time Efficiency-Waste Detection Improvement significantly increases it from 0.754 to 1.42. Regarding the impact of Lean, the findings indicate that the lean variable is not a dominant factor directly affecting the maturity score, but rather plays a pivotal role in shaping the implementation strategy (pp. 9-10). The primary scientific contribution of this study lies in the integration of an interdisciplinary Digital Lean model via the Decision Tree (DT) algorithm to transparently project digital maturity score advancements through targeted intervention paths.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Heribertus Ary Setyadi, Galih Setiawan Nurohim, Wawan Nugroho, Pudji Widodo

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Authors who publish manuscripts through IJCIT agree to the following provisions:
1. The copyright holder of the article is the author.
2. The author grants the right to publish the scientific article to IJCIT as the first publisher. At the same time, the author grants permission/license regarding the Creative Commons Attribution License to other parties to distribute the article.
3. Non-exclusivity matters in the distribution of the Journal, namely the publication of the author's scientific article can be agreed separately (for example, a request to be included in the institutional library or published as a book) then adjust the author as one of the parties and IJCIT as the first publisher.
4. Authors can publish articles online before and during the manuscript submission process (for example, in the Repository or on the website of the organization/institution), as this can encourage the creation of estimates and exchange of citations.
5. Manuscripts and related materials published through this Journal are distributed under the Creative Commons Attribution-ShareAlike 4.0 International License (CC BY-SA).



