Pengembangan Model Kompetisi Antar Platform Di Pasar Ride-Hailing Berbasis Sepeda Motor

Herawatie, Dyah (2025) Pengembangan Model Kompetisi Antar Platform Di Pasar Ride-Hailing Berbasis Sepeda Motor. Doctoral thesis, Institut Teknologi Sepuluh Nopember.

Text 02411760010004-Doctoral.pdf - Accepted Version Restricted to Repository staff only Download (6MB) | Request a copy

Abstract

Dalam beberapa dekade terakhir, bisnis berbasis platform telah menjadi model bisnis yang dominan, termasuk di sektor transportasi. Layanan ride-hailing berbasis sepeda motor di Indonesia merupakan salah satu layanan yang populer di Asia Tenggara. Persaingan antar platform telah memunculkan fenomena multi-homing, yaitu ketika pengemudi atau penumpang menggunakan lebih dari satu platform secara bersamaan. Fenomena ini mempengaruhi keseimbangan pasar dan strategi persaingan. Penelitian ini bertujuan mengembangkan model simulasi menggunakan pendekatan Agent-Based Modeling and Simulation (ABMS) yang merepresentasikan interaksi antara pengemudi, penumpang, dan dua platform ride-hailing yang saling bersaing. Model ini menggambarkan perilaku sistem, termasuk keputusan pengemudi dalam menerima ride request, interaksi dengan penumpang, dan pencocokan permintaan-penawaran dalam jaringan spasial. Data geografis dari wilayah Surabaya Timur digunakan dengan dukungan Geographic Information Systems (GIS) untuk menciptakan lingkungan simulasi yang lebih realistis. Hasil penelitian menunjukkan bahwa pada skenario platform tunggal, strategi Short Distance (SD) unggul dalam waktu tunggu dan dispatch distance, namun produktivitasnya rendah, terutama pada permintaan tinggi. Strategi Long Distance (LD) memberikan pendapatan terbaik pada tingkat permintaan tinggi, meskipun waktu tunggunya lebih lama. Strategi Take-All (TA) unggul dalam berbagai metrik. Mode jelajah random memberikan hasil lebih baik dalam beberapa metrik, tetapi meningkatkan emisi. Radius broadcast optimal berkisar 2-3 km. Dalam skenario kompetisi dua platform, multi-homing meningkatkan pendapatan pengemudi pada tingkat permintaan rendah hingga sedang, serta memperluas akses layanan penumpang, khususnya pada platform kurang dominan. Namun, pada platform dominan, multi-homing dapat meningkatkan waktu tunggu dan menurunkan efisiensi. Pengemudi cenderung single-homing pada permintaan tinggi, bekerja pada platform dominan, berpenghasilan di atas UMR, atau mendapat order di atas 12,5 kali. Sebaliknya, pengemudi lebih memilih melakukan multi-homing atau bekerja lebih dari 8 jam. Penelitian ini memberikan wawasan strategis bagi pengelola platform dalam merespons dinamika pasar. Penelitian lanjutan dapat memperluas kajian dengan mempertimbangkan variasi proporsi pengemudi, market share, distribusi geografis permintaan, strategi harga, serta dampak jangka panjang multi-homing terhadap persaingan dan struktur pasar.
=====================================================================================================================================
In recent decades, platform-based businesses have become the dominant model, including in the transportation sector. Motorcycle-based ride-hailing services in Indonesia are among the most popular mobility-sharing services in Southeast Asia. Competition between platforms has led to the emergence of multi-homing, in which drivers or passengers use more than one platform simultaneously. This phenomenon affects market balance and competitive strategies. This study aims to develop a simulation model using the Agent-Based Modeling and Simulation (ABMS) approach to represent the interactions among drivers, passengers, and two competing ride-hailing platforms. The model captures system behavior, including driver decisions in accepting ride requests, interactions with passengers, and the matching process between supply and demand within a spatial network. Geographical data from East Surabaya, supported by Geographic Information Systems (GIS), creates a more realistic simulation environment. The results show that in a single-platform scenario, the Short-Distance (SD) strategy performs best in terms of waiting time and dispatch distance, but yields lower productivity, especially under high demand. The Long-Distance (LD) strategy provides the highest income during peak demand, albeit with longer waiting times. The Take-All (TA) strategy excels across multiple performance metrics. The random cruising mode improves some metrics but increases emissions. The optimal broadcast radius ranges from 2 to 3 km. In the two-platform competition scenario, multi-homing increases driver income under low to moderate demand levels and expands service access for passengers, particularly on the less dominant platform. However, on the dominant platform, multi-homing may increase waiting times and reduce efficiency. Drivers tend to adopt single-homing when demand is high, when working on a dominant platform, earning above the minimum wage, or receiving more than 12.5 orders. Conversely, drivers are more likely to engage in multi-homing or to work more than 8 hours. This study offers strategic insights for platform managers in responding to market dynamics. Future research may extend the analysis by considering variations in driver proportions, market share, geographic demand distribution, pricing strategies, and the long-term impact of multi-homing on competition and market structure.

Item Type:	Thesis (Doctoral)
Uncontrolled Keywords:	ride-hailing, sepeda motor, ride-hailing berbasis sepeda motor, kompetisi antar platform, multi-homing, Agent-Based Modeling and Simulation (ABMS), Geographic Information Systems (GIS), motorcycle, motorcycle-based ride-hailing, multi-homing
Subjects:	T Technology > T Technology (General) > T57.62 Simulation
Divisions:	Faculty of Industrial Technology and Systems Engineering (INDSYS) > Industrial Engineering > 26001-(S3) PhD Thesis
Depositing User:	Dyah Herawatie
Date Deposited:	05 Aug 2025 08:23
Last Modified:	05 Aug 2025 08:23
URI:	http://repository.its.ac.id/id/eprint/127353

Actions (login required)

View Item