Project Management Services
eBusiness Services E-Business enterprise architecture solutions Business process mapping Project Management Services

Renée Taylor Consulting, Inc.

Enterprise Architecture for Information Systems: Planning and Strategy

Enterprise Architecture Components

Strategic business analysis results in an enterprise strategy, an overview of the organization and its information requirements that is a structured framework for development of integrated applications. There are four major components of an Enterprise Architecture strategy:

  • The Business Direction Model
  • The Information Model
  • The Functional Hierarchy
  • The Application Architecture

The Business Direction Model

The Business Direction Model outlines the goals and objectives of the organization, then further defines critical success factors (CSF's) and key performance indicators (KPI's). Goals are broadly expressed statements that support the mission of the organization. Objectives quantify the goal. Critical success factors are those things that must be in place to achieve the objective and, in turn, to realize the goal. Key performance indicators are the specific measurements of success. An example from the health care industry follows:

Mission: Deliver quality health care at a reasonable cost at a profit.
Goal: Become more cost-effective in acute care facilities.
Objective: Decrease inpatient length of stay (LOS) by 1 day within 6 mos.
CSF: Implement Care Maps for 50% of patient types
KPI: Length of stay

The KPI's are significant because they are quantifiable, can be tracked against a plan, and can provide managers with powerful tools for monitoring the state of the organization. These indicators, however, are rarely required for the day-to-day business functions, and the necessary detail is not generally captured unless mandated by higher levels of management or forced by an application.

To meet executives' needs for management data, it is essential to design applications that show (and perhaps capture) performance indicators. Adaptable models must be built so that current and future management information can be gathered as a natural by-product of normal business procedures without disruption to operational personnel. In this way, the requirement to collect data to monitor the organization's performance can be met without compromising end-user functionality or satisfaction.

The Information Model


The Information Model is a diagram composed of high-level entities and the relationships between them, along with descriptions and examples. It is a model of the information within an organization, and illustrates how that information is linked.

The things of importance (entities) within the business, the properties of those things (attributes), and how they are related (relationships) are depicted on a diagram. In the strategy study, entities are defined at a high level and do not go into specific detail (e.g., address is defined, not street number, street name, city, zip code, etc.).

Later, as more information is gathered during analysis, this model will be amended and further detail added to produce the Business Data Model, which will provide the definition of the structure of all the data that is used or generated. The initial Information Model includes the majority of the entities and relationships that will be in the final, detailed model, with the relationship arcs and all major attributes.

The Information Model is a very powerful communication tool for creating complex models using simple conventions that can be easily explained. This technique is independent of the mechanisms used to hold that information, such as paper forms or computers. As long as the business does not change, the model will remain accurate. Data Model and Entity Relationship Model are synonyms for Information Model, and are often used interchangeably. When the scope of the modeling is across the entire business, the model is called an Enterprise Data Model.

Enterprise data modeling is a technique that aids in:
  • Identifying related and overlapping information across multiple systems elements
  • Reducing data overlaps and inconsistencies and data integrity problems arising as a result
  • Creating architecture solutions for data interfaces and potential integration of data elements that can be mapped to database designs for future in-house Information Technology projects and third-party purchases
  • Developing a direct route to an integrated, flexible information resource based on a clear understanding of information requirements expressed as data structures, interdependencies, and common functionality
  • Planning for high-quality systems that better meet the requirements of business enterprise information processing, derived from business models agreed between managers, users and developers
  • Initiating working practices in the organization to support the implementation of structured techniques for information systems development and the use of supporting CASE tools
  • Improving productivity within Information Technology by clarifying basic concepts at the senior management level and often simplifying requirements or converging redundant development efforts

Enterprise data modeling enables an organization to scope and agree on information requirements for major information systems efforts, such as the acquisition of financial management software, new systems development efforts, and the longer-term Information Technology implementation plan, which will seek to integrate information across multiple current systems.

Enterprise data modeling involves different sets of organizational users, as analysts review data requirements across business areas, determine common patterns of data usage and develop a data model that is representative of the enterprise.

The technique of data modeling identifies valid data entities, relationships, attributes, and unique identifiers in preparation for more flexible, and longer-lasting database designs. Enterprise Architecture, in terms of data modeling, aids in discovering common patterns of data usage, and reduces overlapping current flat file and table elements into a simpler set of standard concepts, converged for better information systems flexibility and integration.
Enterprise Architecture promotes proper selection and adoption of third-party application packages, by offering comparison of business data requirements with package contents. The enterprise approach to data also offers an opportunity to develop standards for data elements and field lengths to ensure future systems compatibility and consistency, improve integrity, and eliminate redundancy in enterprise databases.

The Functional Hierarchy


Function modeling provides a model of what an organization does (its functions) and how those functions can be grouped. The functions within a business are grouped into a hierarchy.

The initial Functional Hierarchy, created during the strategy study, is a list of the major functions of a business. It is created by analyzing the business, using a top-down approach. The starting point and highest function of the hierarchy is the mission statement or goal of the organization. The finishing point is an elementary business function in which a specific unit of work is defined. To be considered elementary, a function must be the smallest unit of work that either retrieves or changes information and that must be either completed or totally undone in order to preserve the integrity of the business information. Like the Information Model, the Functional Hierarchy is concerned with the what of the business — the how is left to system design and individual application design.

The purpose of function modeling is to show how the mission statement is fulfilled by the functions performed within the organization. This is a powerful technique that can lead to fundamental questions about how an organization operates.

The Application Architecture


Once the Information Model and the Functional Hierarchy of the business Enterprise Architecture have been completed, the relationships and dependencies between entities and functions are established through high-level strategic analysis. Every entity must be maintained by at least one function, and every function must use at least one entity.

This interrelationship of entities and functions forms the basis for the application architecture. Entities and functions that are highly dependent on each other are grouped into “building blocks,” e.g., work orders, monitoring, or inspections. Applications, when developed, will consist of one or more building blocks. These are the key to insuring seamless integration of applications. When a “block” is completed, it is available to more than one application, and the blocks need not all be completed at the same time. This allows for intelligent sequencing of development efforts and eliminates redundant development. Perhaps most importantly, it facilitates more accurate cost and schedule estimates, since the blocks identify specific functions (and their complexities) and the entities they use.

Renée Taylor Consulting
Tel. 1 (530) 692 2000
Based in Sacramento
Serving California State government clients, other public sector and corporate clients internationally

expert@rt-consulting.com



Copyright ©2001 Renée Taylor Consulting. All rights reserved.
Web Site Design: Volo Studios, Inc.