IS 341

Business Systems Analysis

Chapter 7 - Notes

Structuring System Requirements: Conceptual Data Modeling

The Second section of the Analysis Phase is Requirements Structuring, which is itself broken down into two parts: Process Modeling and Conceptual Data Modeling. Process flows, decision logic, and data model descriptions of a system must be consistent and complete because each describes different but complementary views of the same system. This chapter deals with Conceptual Data Modeling.

I. Conceptual Data Modeling – a representation of organizational data; shows the definition, structure, and relationships within the data. Purpose: to show as many Rules about the meaning and inter-relationships among the data as possible.

Business Rules – details about the data

Conceptual Data Model – a detailed model that shows the overall structure of organizational data while being independent of any database management system or other implementation considerations.

Data modeling may be the most important part of analysis for four reasons:

1. Characteristics of Data are crucial in designing databases, programs, screens, and reports;

2. Data, rather than processes, are the most complex aspects of most modern systems;

3. The characteristics of Data are permanent (as opposed to reports or programs which are constantly revised over time);

4. Structural information about data is essential to generating programs automatically.

A. The Process of Conceptual Data Modeling – first develop a model of the system being replaced, if one exists (this is essential!), then develop a new one for the proposed new system.

B. Deliverables and Outcomes – the primary deliverable from this section is the Entity-Relationship Diagram (ERD)

***IMPORTANT:

II. Gathering Information for Conceptual Data Modeling – the questions asked must relate to understanding the rules and policies by which the organization operates, i.e., the data model explains what the organization does and what rules govern how work is performed; you do NOT need to know how or when data is processed to do data modeling. ***

A. Top-Down Approach – derives the data model from an intimate understanding of the nature of the business rather than from forms, reports or displays; questions are posed in Business terms.

B. Bottom-Up Approach – information is gathered by reviewing specific business documents – reports, forms, and displays – which appear as data flows on DFDs as the data processed by the system.

III. Introduction to Entity-Relationship Modeling

Entity-Relationship Diagram (ERD) – a detailed, logical, and graphical representation of the entities, associations, and data elements of an organization or business area.

A. Entities – anything about which the organization maintains data; represented by a Rectangle

1. Entity Type (or Class) – a collection of entities that show common properties or characteristics; described only once in the data model

2. Entity Instance – a single occurrence of an entity type/class; occurs as many times as there are entity instances of this entity type

B. Attributes – a named property or characteristic of an entity that is of interest to the organization (NOTE: relationships may also have characteristics; such a relationships is called an Associative Entity)

1. Descriptors – any/every/all attribute(s) are collected for a reason, and therefore provide a description of an entity, therefore ALL attributes are descriptors

2. Identifiers – an attribute which has a unique value for each entity, and therefore provides not only description but also identification; Identifiers are also called Keys

3. (NOTE – NOT IN TEXT) – Domain – the range of values that an attribute may legitimately assume

C. Candidate Keys and Identifiers

Keys – an attribute (or combination of attributes) of an entity which has a Unique value for each entity instance:

1. Candidate Key – any key which may be considered to be used as THE key used for identification within the system

2. Primary Key (Identifier) – the candidate key chosen as the identifier for the system

3. Alternate Key – any candidate key not chosen as the primary key, but which could have been chosen as such

4. Concatenated (Compound/Composite) – key created by combining two or more attributes

5. Foreign Key – a key in one relation (table) used as a non-key attribute in another relation (table) for the purpose of navigating between relations (tables)

Rules for Selecting Keys:

1. The value of the key must not change over the entity’s lifetime

2. The value of the key must be valid

3. The value must not be Null

4. Avoid Intelligent Keys

5. Consider using single-attribute surrogate keys instead of large composite keys (create a GAME_ID key instead of HOME_TEAM-VISITING_TEAM-DATE)

D. Multi-valued Attributes – an attribute that may take on more than one value for each entity instance (this takes us to the first step of Normalization – covered in Chapter 9)

1. Strong (independent) Entity - one that exists regardless of the existence of any other Entity; its primary key contains no attributes making a dependent on any other Entity

2. Weak (dependent/attributive) Entity - and entity whose existence depends on the existence of other entities

E. Relationships – associations between the instances of one or more entity types

IV. Conceptual Data Modeling and the E-R Model

*****Since the largest expense of any system is the maintenance of the system, we gain efficiency (lower costs) by maintaining systems at the Rule Level (the data level, the repository) rather than at the Code Level (the processes, the programs).

A. Degree of a Relationship – the number of entities participating in a given relationship

1. Unary Relationship -

a. Employee - Manages - Employee

b. Person - Is Married To - Person

2. Binary Relationship

a. Employee – Is Assigned – Parking Place

b. Product Line – Contains - Product

c. Student – Registers For - Course

3. Ternary Relationship

Entities: Vendor, Part, and Warehouse

Relationship: Ships (an Associative Entity)

Attribute of Relationship: Quantity

The relationship Ships keeps track of the Quantity of a given Part that is Shipped from Vendor to Warehouse.

B. Cardinalities in Relationships – the number of instances of one entity instance that can (or must) be associated with each instance of another entity

1. Minimum and Maximum Cardinalities – minimum number of instances of entity B that can/must be associated with each instance of entity A.

A Distributor must order a minimum of 6 of an item from the Manufacturer, so the Minimum Cardinality on an order is 6 for an Order to exist. The Manufacturer has a limit of 4 Vendors for each item, so the Maximum Cardinality for an Item Venbdor would be 4.

2. Associative Entities - an entity that inherits its primary key from more than one other Entity (parents)&/or comes about due to 1:Many or Many:Many relationships; each part of that concatenated key points to only one instance of each of the connecting entities; this happens when a Relationship has attributes which cannot be expressed as relating to either entity alone (e.g., a Grade for a Student in a given Class), so the Relationship becomes a kind of an entity by virtue of having attributes. An Associative Entity may actually be a matter of your perspective of the data model – two persons may describe it differently: one as an Entity, one as an Associative Entity.

C. (NOTE – NOT IN TEXT) – Views – a particular way of looking at a system design

1. Schema – the Physical view (Computer’s view) of the system; there is only one (1) schema for any system. The computer is “God”, as it were, and knows everything about the entire system, what parts/pieces there are, where they are located, and how they are connected.

2. Subschema – the Logical view (User’s view) of the system; there may be one (1) or many (more than one) subschemas for a given system. The user sees only that part of the system that is relevant to him/her, and has no knowledge of (or access to, usually) those parts of the system unrelated to him/her.

V. An Example of Conceptual Data Modeling at Hoosier Burger

VI. Pine Valley Furniture’s WebStore: Conceptual Data Modeling

VII. Selecting the Best Alternative Design Strategy - Two steps:

1. Generate as Comprehensive set of alternative design strategies (what is Comprehensive? How can we accommodate ALL possible alternatives?);

2. Select the one most likely to result in the desired information system given all the organizational/economic/technical constraints (Just what is “the desired information system”? Desired by whom? Which requirements are filled and which unfilled?)

Design Strategy – a particular approach to developing the system

A. The Process of Selecting the Best Alternative Design Strategy:

1. Divide requirements into different system capabilities from bare minimum to most elaborate/advanced; or, alternatively, represent different positions/requirements of different, conflicting parts of the organization;

2. List the different potential implementation environments (hardware, software, networks) that could be used to deliver different sets of capabilities (choices here may limit development options in the future! It’s NEVER a good idea to choose physical models first because that limits the system’s capabilities fro the start!)

3. Propose different ways to source/acquire the various sets of capabilities for the different implementation environments.

Analysts and Management (Steering Committee) may disagree on what is “Best” – get used to it! The project may very well be stopped altogether at this point due to considerations of costs/risks vs. benefits, changing organizational needs, changing priorities of other projects, or resource constraints.

Ok, GET REAL!!!!! In most cases it is tough enough to generate one acceptable, comprehensive plan that covers all needs/requirements and is within organizational constraints. Three really good alternatives? GOOD LUCK!!!!!

Many organizations do not have the time/funding/(enter your own favorite constraint) to allow generation of multiple alternatives. Most of the time it is the job of the analysts to come up with the optimal solution at the first attempt, and it better be the right one!

B. Generating Alternative Design Strategies – many analysts can see the best solution as an obvious choice, but this reflects experience in the field and with the organization. The reason for suggesting three possible solutions is to advance low/medium/high dollar costs to give the steering committee a choice. However, all must consider:

1. Minimum system requirements and mandatory features, such as 1. Data kept in files; 2. System outputs; 3. Analyses to generate the information in the system outputs; and Expectations of accessibility, response time, or turnaround time for system functions.

2. Constraints on system development such as 1. a Date when the new system is required; 2. available financial and human resources; 3. Elements of the current systems that cannot be changed; 4. Importance/Dynamics of the problem that may limit how the system can be acquired.

VIII. Developing Design Strategies for Hoosier Burger’s New Inventory Control System

A. Selecting the most Likely Alternative (for Hoosier Burger)