Introducing the Enterprise Management Framework

by Andre Milchman
Enterprise Architecture is NOT an Information Technology issue... it is a Management issue.
- John A. Zachman

The Enterprise Management Framework (EMF) attempts to formalize, conceptualize and articulate principles related to the organization of the Enterprise in a simple and coherent framework and to apply this framework to enterprise engineering. The model intentionally liberates itself from the constraints imposed by Industrial Age management standards and adds principles focused on designing the Digital Enterprise.

Definition
The Digital Enterprise is a form of organization used by individuals to perform purposeful computational activities.

The EMF covers the following aspects of enterprise management:

Figure 1.1. Aspects of the Enterprise Management Framework.

1. Enterprise Computation

Principle 1.1. (Computation)
All enterprise activities which manipulate digital content are considered to be computational.

This principle means that any business, social, scientific, artistic, engineering, or design activities that deal with (create/modify/destroy/display) digital records, files, documents, software, video, audio, images, messages, or any other type of digital content are categorized as computation.

The EMF distinguishes computational activities from physical/chemical enterprise activities in order to highlight the need for designing new management models, adapted to the requirements of a modern-day digital enterprise.

Principle 1.2. (Computational Agents)
Computational activities are performed by both human and artificial agents.

In the case of artificial agents, computation is performed by software applications running on computer hardware. Similarly, one can imagine human mind as "software" that runs on the "hardware" of the brain.

Principle 1.3. (Tools and Personal Assistants)
To successfully perform computational tasks, human agents use digital tools and virtual assistants that run on devices with input and output capabilities.

Although the situation may change with the advent of brain–computer interfaces, modern-day humans cannot directly manipulate digital content and therefore rely on digital tools—desktop, web, and mobile applications—or digital personal assistants.

Principle 1.4. (Platforms and Infrastructure)
The entire lifecycle of artificial computational agents is managed by digital platforms which run on digital infrastructure.

Today, Cloud Computing technologies provide the means for efficient and effective management, maintenance, and provision of infrastructure and platforms.

2. Enterprise Composition

Principle 2.1. (Composite Agents)
The Enterprise creates composite agents—units—that consist of individual (human and artificial) agents.

Units create synergy between individual human and artificial agents in order to increase (1) intelligence (the ability to process information), (2) purposefulness (the ability to discover and satisfy the needs of others), (3) robustness (the ability to provide high quality of service), (4) autonomy (the ability to make decisions), (5) governability (the ability to be subjected to authority), (6) interactivity (the ability to collaborate, cooperate, and communicate with others), (7) adaptivity (the ability to learn and evolve), and (8) resilience (the ability to recover from errors).

Principle 2.2. (Levels of Scale )
The Enterprise has two levels of scale: the macro (enterprise) level is implemented as a network of units; the micro (unit) level is implemented as a network of individual agents.

When constructing artificial agents, we can discover other levels of scale—agent level, component level, and object level—however, those are the concerns of software design, not business design.

Principle 2.3. (Enterprise-Level Business Blocks)
Only two types of units—functional units and executive units—can be used for constructing the enterprise level structure.

Functional units perform core activities that directly relate to the purpose of the Enterprise, whereas executive units ensure that the units under their authority contribute their best to the success of the whole.

Figure 2.1. The Enterprise is a network of executive and functional units.
Principle 2.4. (Exclusivity of Composite Agents)
Only composite agents (units) can be used for constructing the enterprise level structure. Individual agents must become parts of units.

Although most modern-day organization are complex hierarchies with more than two levels of scale and have human agents (SVP and above) at the top levels, in the 21st century this should be considered as a business design antipattern. In the age of hyper-fast, hyper-volume, and hyper-complexity information streams, human agents are not capable of (1) effective and efficient information processing, (2) making, exposing, recording, and learning from decisions, and (3) providing the desired quality of service. Therefore, they should become parts of stronger enterprise centers—units.

Principle 2.5. (Unit-Level Business Blocks)
Individual (human and artificial) agents are the main building blocks used for constructing the unit level. Digital tools and digital personal assistants are considered to be the supporting building blocks that complement the capabilities of the individual agents.

Figure 2.2. A unit is a network of individual agents.
Principle 2.6. (Aspects of Unit Design)
The Unit is defined by its Purpose, Function, Process, Structure, and Culture.

Purpose connects the unit to its customers. Function enables the customers to interact with the unit. Process is characterized by a set of operations that support the unit's Function. Structure is represented as a network of agents that play certain roles and perform various tasks within the Process. Culture is a collection of values, beliefs, and behaviors that energize the Structure.

Principle 2.7. (Unit Design Principles)
Units must be designed to satisfy the requirements of unity, coherence, completeness, and cohesion.

Unity determines the unit's effectiveness—the ability to focus efforts toward a unifying purpose. Coherence means that the unit's Purpose, Function, Process, Structure, and Culture logically follow each other and create a unified and balanced whole. Cohesion ensures that every part of the unit is clearly linked to the next and establishes continuity within or towards other units. Completeness means that all digital constructs clearly and thoroughly support the unit's Purpose, Function, Process, Structure, and Culture.

3. Enterprise Interchange

Principle 3.1. (Relations)
Enterprise units are connected with one another and with the environment through different types of relations: hierarchical (asymmetric: directing/reporting) relations, collaborative (sharing) relations, exchange (buy/sell, give/take, etc.) relations, and other types of relations.

Hierarchical relations are established between executive units on one side, and units under their authority (which could be either executive or functional) on another side. Collaborative relations can be established between all enterprise units who share information with each other, often by broadcasting. Exchange relations are typically established between enterprise units and the environment.

Principle 3.2. (Thick Boundaries)
Enterprise units enclose themselves in digital boundaries that enable and govern their relations with other units and the environment and protect them from harmful, inappropriate, and undesirable interactions.

Austrian architect Christopher Alexander, author of The Pattern Language remarked that a thick boundary is a kind of interchange zone. Where two zones, lying side by side, have their own integrity, it is not enough to have a sharp edge to separate them. In many cases, and especially in cases of living systems, there is a need for exchange, interchange, flows passing across the boundary, subtle kinds of filtering, places where interaction of the elements coming from either side can take place. Alexander believed that in general it is necessary to think of boundaries as very large. In The Nature of Order: The Phenomenon of Life he wrote, If the boundary is very much smaller than the thing being bounded, it can't do much to hold in or form the center. A two-inch border cannot hold a three-foot field. In a room, the boundary between floor and wall needs more than a six-inch molding - a wainscoting, 30 inches high, is more in scale with both. An effective boundary for the river Seine consists of roads, walls, paths, quays, trees, something almost as massive as the river itself.

Principle 3.3. (Facades)
A unit's boundary consists of facades of different orientation: external, internal, and private.

External facades of all enterprise units combined represent the public facade of the Enterprise, whereas internal facades combined represent the private facade of the Enterprise.

Principle 3.4. (Interfaces)
A unit's facade consists of interfaces that represent points of interactions between the unit and the environment.

These interfaces can be realized as web portals or applications, mobile apps, microservices or APIs, etc.

Principle 3.5. (Artifacts)
Interfaces are used to deliver services, products, by-products, or notifications from providers to consumers.

Enterprise units produce combinations of services, products, or by-products. A service is an immaterial exchange of value that requires specific input from another entity and contributes to the achievement of its mission. A product is any digital content that is produced as a result of some process without a specific input of any particular entity. A by-product is any digital content that is produced unintentionally as a result of other process.

4. Enterprise Coordination

Principle 4.1. (Enterprise-Level Coordination)
At the enterprise level, units coordinate their activities using expectation-based, event-driven coordination mechanisms.

Traditional coordination mechanism, business process, has certain shortcomings if used for knowledge work activities. With the excessive focus on design-time activities (such as process modeling and simulation), Business Process Management (BPM) has not fulfilled the promise of agility, flexibility, and efficiency and was unable to achieve the desired levels of maturity in standardization and commercial applications. Worknet, the next generation coordination mechanism, breaks the traditional command-and-control model of management and shifts focus from design-time to run-time activities, which greatly improves agility and flexibility of knowledge workflows.

Principle 4.2. (Unit-Level Coordination)
At the unit level, individual agents coordinate their activities using either command-and-control or expectation-based, event-driven coordination mechanisms.

Command-and-control coordination is critical to the success of highly standardized production and administrative workflows. Expectation-based, event-driven coordination is more suitable for knowledge-based and relationship-based workflows.

5. Enterprise Communication

Principle 5.1. (Enterprise-Level Communication)
At the enterprise level, intermediation agents enable asynchronous communication based on queueing and publish/subscribe mechanisms, whereas mediation agents enable location transparency, protocol transformation, message enhancement, and quality of service.

Asynchronous communication combined with standard mediation services minimizes interdependence and simplifies the integration of enterprise units.

Principle 5.2. (Unit-Level Communication)
At the unit level, both synchronous and asynchronous communication styles are used; the former supports command-and control coordination; the latter, expectation-based, event-driven coordination.

While asynchronous communication increases autonomy, agility, and adaptability of enterprise agents, synchronous communication, which is simpler, faster and easier to use, will remain the preferred choice for many enterprise scenarios.

6. Enterprise Transformation

Principle 6.1. (Phases of Transformation)
Enterprise transformation is executed in four phases: (1) social transformation, (2) systemic transformation, (3) technological transformation, and (4) unit transformation.

The first phase partitions the Enterprise into units capable of performing complex computational activities at desired service levels. The main challenge of this phase lies in eliminating human agents from the enterprise level, which might be perceived as loss of status by many. The second phase establishes thick boundaries around enterprise units in order to formalize their interactions with the environment. The third phase implements new coordination and communication mechanisms that minimize operational interdependence between units. Whereas the first three phases focus on the enterprise-level transformation, the fourth phase triggers the transformation of individual units, which can progress at an individual pace and rhythm.