Analisis Dampak Intensitas Penggunaan AI Terhadap Produktivitas Karyawan: Peran Mediasi Perubahan Cara Kerja dan Kepercayaan Pada AI

Jafarian, Ahmad Reza (2026) Analisis Dampak Intensitas Penggunaan AI Terhadap Produktivitas Karyawan: Peran Mediasi Perubahan Cara Kerja dan Kepercayaan Pada AI. Masters thesis, Intitut Teknologi Sepuluh Nopember.

Text 6032242011-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (2MB) | Request a copy

Abstract

Transformasi digital mendorong adopsi teknologi Artificial Intelligence (AI) seperti Large Language Models (LLM) dan Generative AI di berbagai perusahaan, termasuk perusahaan yang bergerak di sektor jasa. Transformasi yang terjadi merupakan bentuk upaya untuk meningkatkan efisiensi, akurasi, dan produktivitas sumber daya manusia. Salah satu perusahaan jasa logistik di Jakarta Barat telah mengimplementasikan teknologi tersebut untuk meningkatkan produktivitas karyawan di sana. Penelitian ini difokuskan pada analisis dampak intensitas penggunaan AI terhadap produktivitas karyawan office di beberapa anak perusahaan dalam satu grup usaha jasa logistik. Adapun kondisi yang ingin dianalisis adalah dampak intensitas penggunaan AI terhadap produktivitas melalui variabel mediasi perubahan cara kerja dan kepercayaan pada AI. Penelitian menggunakan pendekatan kuantitatif eksplanatori dengan desain survei cross-sectional. Pengumpulan data dilakukan melalui kuesioner digital yang disebarkan kepada karyawan office di perusahaan terkait yang aktif menggunakan AI. Proses analisis data dilakukan menggunakan Structural Equation Modeling (SEM) untuk menguji hubungan kausal serta efek mediasi antarvariabel. Hasil penelitian menunjukkan bahwa intensitas penggunaan AI tidak berpengaruh langsung secara signifikan terhadap produktivitas karyawan (koefisien jalur 0,106; p-value 0,059), melainkan bekerja sepenuhnya melalui mekanisme mediasi (full mediation). Perubahan cara kerja terbukti memediasi hubungan tersebut secara signifikan (pengaruh tidak langsung 0,164; p-value 0,000), demikian pula kepercayaan pada AI yang merupakan jalur mediasi paling dominan (pengaruh tidak langsung 0,258; p-value 0,000). Secara keseluruhan, model mampu menjelaskan 60,4% varians produktivitas (R² = 0,604) dengan tingkat kesesuaian model yang baik (SRMR = 0,078; GoF = 0,519). Temuan ini menegaskan bahwa peningkatan produktivitas tidak ditentukan oleh seberapa sering AI digunakan, melainkan oleh terbangunnya kepercayaan terhadap AI dan terjadinya transformasi cara kerja sebagai jembatan utama antara intensitas penggunaan AI dan produktivitas karyawan.
====================================================================================================================================
Digital transformation has accelerated the adoption of Artificial Intelligence (AI) technologies, such as Large Language Models (LLMs) and Generative AI, across various companies, including those operating in the service sector. This transformation represents an effort to enhance efficiency, accuracy, and human resource productivity. One logistics service company in West Jakarta has implemented these technologies to improve employee productivity. This study focuses on analyzing the effect of AI usage intensity on the productivity of office employees across several subsidiaries within a logistics service group. Specifically, it examines the effect of AI usage intensity on employee productivity through the mediating variables of work pattern changes and trust in AI. The study employs a quantitative explanatory approach with a cross-sectional survey design. Data were collected through digital questionnaires distributed to office employees who actively use AI within the selected companies. The data were analyzed using Structural Equation Modeling (SEM) to examine causal relationships and mediation effects among the variables. The findings reveal that AI usage intensity has no significant direct effect on employee productivity (path coefficient = 0.106; p-value = 0.059). Instead, its effect operates entirely through mediation (full mediation). Work pattern changes significantly mediate this relationship (indirect effect = 0.164; p-value = 0.000), as does trust in AI, which represents the most dominant mediating pathway (indirect effect = 0.258; p-value = 0.000). Overall, the model explains 60.4% of the variance in employee productivity (R² = 0.604) and demonstrates a good model fit (SRMR = 0.078; GoF = 0.519). These findings confirm that improvements in employee productivity are not determined by how frequently AI is used, but rather by the development of trust in AI and the transformation of work patterns, which serve as the primary mechanisms linking AI usage intensity to employee productivity.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	Artificial Intelligence, Produktivitas Karyawan, Perubahan Cara Kerja, Kepercayaan terhadap AI, Structural Equation Modeling. ============================================================ Artificial Intelligence, Employee Productivity, Work Pattern Changes, Trust in AI, Structural Equation Modeling
Subjects:	H Social Sciences > HD Industries. Land use. Labor > HD56.25 Industrial efficiency--Measurement. Industrial productivity--Measurement.
Divisions:	Faculty of Business and Management Technology > Management Technology > 61101-(S2) Master Thesis
Depositing User:	Ahmad Reza Ja`farian
Date Deposited:	28 Jul 2026 04:14
Last Modified:	28 Jul 2026 04:14
URI:	http://repository.its.ac.id/id/eprint/138374

Actions (login required)

View Item