Reinterpreting the Unified Process in the AI Era

ASAMI, Tomoharu Created: 2026-03-16

In the previous article, we organized the development process for the AI era by positioning the Unified Process (UP) as the structural backbone of the process and Component-Based Development (CBD) as the central structure of development.

The Unified Process defines the software development process through three core principles: Iterative & Incremental development, Architecture-Centric design, and Use-Case Driven development.

These principles remain valid even in the age of AI.

However, in an environment where AI-based code generation has become commonplace, the meaning and role of each principle need to be understood somewhat differently from how they were interpreted in the past.

In this article, we revisit the three fundamental principles of the Unified Process as a guide and re-examine the nature of the development process in the AI era.

Three Principles of UP in the AI Era

The Unified Process presents three fundamental principles that support the software development process.

  • Architecture-Centric

  • Use-Case Driven

  • Iterative & Incremental

These principles served as important guidelines for stabilizing large-scale software development at the time when the Unified Process was proposed.

Even in the AI era, these principles themselves do not become obsolete.

Rather, in an environment where AI-driven code generation accelerates development, these principles may become even more important.

Let us reinterpret the meaning of each principle within development environments that assume the presence of AI.

Meaning of Architecture-Centric

In the Unified Process, the Architecture-Centric principle means establishing the system architecture early and advancing development around that architecture.

The architecture defines the core structure of the system and stabilizes the direction of development.

This idea remains important even in the AI era.

In fact, as AI-driven code generation becomes easier, the role of architecture in stabilizing the overall system structure becomes even more important.

AI can generate individual pieces of code rapidly, but it is not particularly good at autonomously maintaining the structure of an entire system.

In development for the AI era, the productivity of implementation generation is strongly influenced by the quality of the structure established during the refinement phase.

Therefore, the structure and boundaries of the system must be designed by humans.

In this sense, being architecture-centric in the AI era can be understood as establishing structural constraints that control AI-generated implementations.

Meaning of Use-Case Driven

In the Unified Process, the Use-Case Driven principle means organizing requirements around use cases and using them as the starting point for design and implementation.

A use case is one of the key representations of requirements, describing how a system will be used.

In the AI era, the role of use cases becomes even more important.

This is because AI systems can now generate specifications and code from structured natural language descriptions such as use case narratives.

In other words, use cases are no longer just models for documenting requirements; they function as specifications that explain system behavior to AI.

In this sense, being use-case driven in the AI era can be understood as advancing development around executable specifications.

Meaning of Iterative and Incremental

In the Unified Process, iterative and incremental development refers to the idea of developing software through repeated cycles while gradually growing the system step by step.

Software development is an activity with a high degree of uncertainty, and it is difficult to create a complete design from the beginning.

For this reason, it is important to gradually improve the system by repeating short development cycles.

In the Unified Process, development progresses through multiple iterations.

In each iteration, a series of activities such as the following are performed.

  • Business

  • Requirements

  • Analysis

  • Design

  • Implementation

  • Testing

In other words, each iteration functions as a small development cycle.

These activities are not limited to a specific phase but are repeated throughout all phases.

However, the emphasis of activities differs depending on the phase.

For example, in the Inception phase the focus is on understanding the problem domain and organizing requirements, in the Elaboration phase the focus is on establishing analysis models and system structure, and in the Construction phase the focus is on implementing functionality.

Thus, in the Unified Process, iterations create the rhythm of development while phases adjust the focus of development.

Furthermore, through the concept of incremental development, each iteration produces a working system at its completion.

In other words, each iteration is not merely a boundary of work but a unit through which the working software grows step by step.

Ensuring that a working system is produced at the end of each iteration is an important principle of the Unified Process.

Even in the AI era, this fundamental idea remains unchanged.

However, in AI-assisted development the speed of iterations increases dramatically.

Because AI can generate implementation code rapidly, adjustments among specification, analysis, design, and implementation can be performed in much shorter iteration cycles.

Therefore, iterative and incremental development in the AI era can be understood as a development style in which specification, analysis, design, and implementation are repeatedly refined through rapid cycles of interaction with AI.

Phase Structure Reinterpreted in the AI Era

The Unified Process organizes the development process into four phases.

These phases are defined as a framework for organizing the progression of a project.

Even in the AI era, this fundamental structure remains valid.

However, in development environments where AI-based code generation has become commonplace, the role of each phase can be interpreted somewhat differently.

Inception Phase

In the Inception phase, the objectives and scope of the project are clarified.

At this stage, the main activities involve understanding the problem domain and organizing the fundamental requirements.

Even in the AI era, the role of this phase does not change significantly.

However, because the knowledge and concepts organized at this stage form the basis of later models and specifications, the clarification of terminology and concepts becomes even more important.

Elaboration Phase

In the Elaboration phase, the architecture and fundamental structure of the system are established.

In traditional development, the main objective of this phase was to determine the direction of the design and verify high-risk elements.

In the AI era, the importance of this phase becomes even greater.

During Elaboration, the fundamental structure of the system is established, including the domain model, component structure, service structure, DSL, and execution platform.

This structure becomes the framework upon which AI generates implementation code.

In this sense, Elaboration in the AI era can be understood as the phase in which a system structure understandable to AI is established.

Construction Phase

In the Construction phase, the system’s functionality is implemented.

In traditional development, the central activity of this phase was writing program code.

However, in environments where AI can generate code, the nature of this phase changes significantly.

When the system structure has been established during Elaboration, AI can generate implementation code based on use cases.

Therefore, in the AI era the Construction phase centers on activities such as writing use cases, refining specifications, and reviewing generated code.

In other words, Construction can be understood not as the phase of writing code, but as the phase in which implementations are derived from use cases.

Transition Phase

In the Transition phase, the system is moved into the actual operational environment.

Activities in this phase include deployment, operational adjustments, and the collection of feedback.

Even in the AI era, the fundamental role of this phase remains largely unchanged.

However, because AI makes implementation changes easier, feedback obtained from operations can be reflected more rapidly in the next iteration.

From this perspective, the three principles of UP define the fundamental principles of development, while the phase structure organizes the progression of a project.

Next, we consider what kind of technical foundation can realize these principles and processes in development environments of the AI era.

SimpleModeling: A Development System in the AI Era

As discussed above, the three principles proposed by the Unified Process—Architecture-Centric, Use-Case Driven, and Iterative & Incremental—remain valid development principles even in the age of AI.

However, in environments where AI-based code generation has become common, a coherent technological and methodological framework is required in order to realize these principles in actual development.

SimpleModeling can be positioned as such a development system.

In SimpleModeling, knowledge organization and execution platforms are connected through a layered structure such as the following.

BoK (Body of Knowledge) Concept Model Analysis Model DSL (Cozy) Execution Platform (CNCF)

In this structure, humans design the conceptual structure, boundaries, and architecture of the system, while AI performs rapid transformations from specification to implementation within that structure.

As a result, a development style emerges in which implementations are derived from use-case-centered specifications.

In this sense, SimpleModeling can be understood as a development system that integrates the reinterpreted Unified Process of the AI era with Component-Based Development.

Summary

In this article, we reexamined the meaning of the development process in the AI era through the three core principles of the Unified Process: Architecture-Centric design, Use-Case Driven development, and Iterative & Incremental development.

In environments where AI-based code generation is common, the center of development shifts from writing code to designing system structure and organizing specifications presented to AI.

In this context, architecture functions as the structure that constrains AI generation, use cases function as specifications that guide implementation, and iterations function as rapid development cycles driven through interaction with AI.

The principles proposed by the Unified Process therefore continue to function as important guidelines for stabilizing software development in the AI era.

SimpleModeling can be positioned as a development system that implements these principles within practical development environments.

References

Glossary

UP (Unified Process)

A process model based on UML, characterized by iterative, use-case–driven, and architecture-centric development. It has derivatives such as the Rational Unified Process (RUP) and provides a foundation for practicing Component-Based Development (CBD).

CBD (Component-Based Development)

Component-Based Development (CBD) is a software development approach in which systems are constructed and reused through components that define clear responsibilities, contracts, and interfaces. Components are designed to be independent and replaceable, enabling loosely coupled architectures that improve maintainability and reusability. In the logical model, a component serves as an abstract structural unit defining functionality and contracts; in the physical model, it corresponds to implementation and deployment units.

Development Process

A development process is the entire set of activities involved in building, deploying, and maintaining software systems. It encompasses planning, modeling, implementation, testing, and delivery as a systematic flow of development activities.

Elaboration Phase

The phase in which the system’s structural backbone is established by advancing analysis and design based on the requirements and directions defined in the inception phase. In the AI era, its most critical role is to refine context, boundaries, and assumptions, eliminating ambiguity and contradiction, and to establish the architecture baseline referenced by the AI.

Activity

A concrete action or task performed within an Activity Space to move an Alpha to a more advanced state, typically producing or refining Work Products.

Construction Phase

The phase in which the system is implemented at scale through iterations, based on the architecture baseline established during the elaboration phase. In the AI era, this phase refers to the stage where AI becomes the primary implementation agent, generating large volumes of code and tests with consistent quality under given structures and constraints.

Inception Phase

The phase in which the project’s goals, problem domain, and scope are defined, and the value and feasibility to be validated are clarified. In the AI era, its central role shifts from fixing detailed specifications to defining the outline of the context shared between humans and AI.

Transition Phase

The phase in which the developed software is transitioned into actual operation and delivered to users. In the AI era, it is positioned as a context-update phase that captures insights from usage and operation and feeds them into the next inception phase.

DSL (Domain Specific Language)

A DSL (Domain-Specific Language) is a language designed for a particular domain, enabling direct and concise expression of the domain’s concepts and structures. Compared to general-purpose programming languages (GPLs), DSLs offer a higher level of abstraction tailored for domain-specific problem solving and automation.