IM TAXI – AI-Driven ERP for Fleet Optimization
IM TAXI, a traditional taxi company, aimed to transition into digital operations by starting from the core system, not consumer apps. With only GPS trackers and mileage meters installed in vehicles, the company lacked centralized data, analytics, and decision support. This project focused on building an ERP with AI to maximize fleet utilization, identify traffic demand patterns, and improve driver and vehicle allocation through data-driven insights.

Product Manager
6 months
3 Frontend Engineers,
3 Backend Engineers,
1 Quality Assurance,
1 System Tester,
2 UX Designer
Project Value
IDR 550 - 750 million (2025)
30+
Overview
IM TAXI is a conventional, enterprise-scale taxi company in Japan. Prior to this project, their operations relied mainly on hardware-level data such as GPS trackers and rate/mileage counters installed in each vehicle. While these tools captured raw data, they were not connected to a centralized system that could support operational planning, financial reporting, or customer analysis.
The company had no CRM to identify loyal customers or design targeted campaigns, limited visibility into fleet utilization patterns, and relied heavily on manual processes for financial reporting. Driver and fleet allocation decisions were made based on experience and intuition rather than historical or predictive data.
The goal of this project was to build an AI-powered ERP that unifies operational, financial, and analytics data—enabling management to optimize fleet deployment, reduce manual work, and support long-term digital transformation.
My Contribution
This was my first project in the mobility and transportation domain, and I worked as a solo Product Manager, owning the product end-to-end.
I led the product discovery, system design, PRD creation, stakeholder alignment, and iteration cycles. One of the biggest challenges was understanding user behavior and decision-making patterns in Japan, which differ significantly from Indonesia—especially for drivers, operations teams, and leadership.
Rather than starting from a driver-facing app, I focused on building a strong operational backbone: an ERP capable of turning raw vehicle data into insights that could guide daily allocation, financial tracking, and strategic decisions.
Problem & Opportunity
Identifying the Core Problems
IM TAXI faced several critical challenges:
No CRM
The company could not identify loyal customers or design targeted promotions and campaigns.Inefficient fleet allocation
Drivers were assigned without clear insight into high-demand areas by date and time, leading to suboptimal utilization.Manual financial reporting
Finance teams relied on manual consolidation, making reporting slow and error-prone.
Despite collecting large volumes of vehicle data, none of it was used to predict demand, optimize allocation, or support planning.
This created a clear opportunity to transform raw data into operational intelligence.
Proposed Solution: AI-Driven ERP for Fleet & Financial Intelligence
I positioned the solution as an ERP-first system, not a driver app, with AI as a decision-support layer.
Key principles:
Centralize all operational and financial data
Use AI to support—not override—human decision-making
Focus on fleet utilization and financial clarity
Core capabilities included:
Fleet utilization analytics to understand demand patterns
AI-assisted area and time-based allocation of drivers and vehicles
CRM module to identify loyal customers and support campaigns
Financial module to reduce manual reporting
Management dashboards for operational and performance visibility

Ideation & Validation Process
Working Closely with Leadership
I iterated directly with BOD-level stakeholders, refining priorities based on strategic goals rather than surface-level features.
Designing for Cultural Context
User behavior in Japan required careful consideration:
Clear accountability
Predictable workflows
High trust in system accuracy
I validated concepts through continuous iteration, ensuring the system aligned with both operational reality and cultural expectations.


The Result
The ERP enabled IM TAXI to:
Reduce manual operational and financial work
Gain clearer visibility into fleet utilization patterns
Support better allocation decisions using data
Establish a foundation for future digital initiatives
The system shifted operations from intuition-based decisions to structured, data-supported planning.
Challenges & Learnings
Mobility Systems Are Deeply Operational
Unlike typical digital products, mobility systems interact directly with real-world constraints—time, location, and human behavior.
Cultural Understanding Is as Important as Technology
Success required adapting workflows to Japanese working styles and expectations, not simply reusing familiar patterns.

Key Takeaways
Digital transformation can start from systems, not apps
AI delivers value when grounded in real operational data
Fleet optimization requires both analytics and human judgment
A strong ERP foundation enables long-term scalability
Get Athos Pro

