Fall 2015 Information System 6840


Computer applications in the non-traditional methodologies have put a requirement on data modeling. Information in the real world is often vague and currently more effort is being put in database design. Record keeping is becoming an important aspect of all businesses. In the information society, it has become increasingly important to maintain and use databases in businesses. All kind of data ranging from contact and email information to record of sales and financial information are stored in databases. Databases require a data model that refers to a conceptual representation of the data structures such as data objects, and the rules that govern the objects operations. Data model focuses on the type of data required and the way it should be organized (Yan and Ma 1881)... It focuses on representing the business data as the user view it in it in reality. Data modeling shows the structure and relationship within data. It is the most crucial part of the information system requirement. The type of data captured during data modeling plays a fundamental role in the design of databases printed reports, computer screens and programs. Due to its ability to organize and structure data, data modeling is widely used in industries. Modeling and analysis are crucial because it enables businesses to identify their needs and problems and develop potential solutions. The data model can be a physical model, logical model or conceptual model. In the real business world, modeling goal keeps changing and thus the role of modeling has become crucial particularly in the early designing phase (Calì, Gottlob and Pieris 1888). Other benefits of the data model are communication and precision during business processes. The purpose of this manuscript is to discuss the role of data modeling in system analysis. It will focus on what system analysis, data model, and the importance of data modeling in system analysis and how it develops in system analysis.

System analysis

System analysis refers to a process of gathering data, understanding the process involves in data collection, identifying the problems associated with it and providing suggestion on how to improve the process. It involves the study of gathering operational data, business processes, understanding the flow of information and evolving a solution for overcoming the existing problems in the system (Muller 113). The function of system analysis is to provide answers to all questions related to business processes. System analysis is crucial in any organizations because it offers a technique for handling complex problems that can hinder efficient resources allocation (Mallikaarachch).

Data modeling

Data model provides a description of how different type of data should be used to meet the need of the user. It is a set of concepts applied that can be applied by an organization to describe the operation and structure of the database. The structure of data refers to the type of data, relationship, constraints that define the database template (Dixit and Kumar 189). A data model provides operations in the database that allow both retrieval and update of organizational data. It is a conceptual representation of a particular system of an organizational information requirement. A conceptual data model represents organizational data, and its purpose is to show the rules about the interrelationship and meaning of the data. The information regarding the processes of a business for the development of high-level data model, and its explanation is collected through gathering information (Calì, Gottlob and Pieris 1888). Data modeling provides an effective method for organizing and structuring data that deals with the database management system. The management of both large structured and unstructured data is carried out by the use of information systems. The function of the data model is to describe the structured data for storage in databases. The structural information about organizational data is essential for the generation of the automatic program. Typically, it does not describe the unstructured data. It documents the database and file requirements for an information system. Data model is a powerful communication tool for users, and it can also be applied in businesses as a blueprint for the data system that is created at different levels of details. Data model in capacity acts as a bridge between the database storing critical and relevant data content to the real work information (Chan 59). Blueprint model presents many benefits to the data modeling. It data model known by very few people outside the information technology and it has many parallels familiar with many people. One of the most common blueprints parallels is that it translates very complicated and technical business undertakings into a set of visual diagrams that easily understand by not only the experts but also layperson. One of the most important goals of data modeling is to help the user and developer to understand the information requirements. It is very effective in communicating and expressing the requirements of a business. Due to differences in companies, data modeling differs significantly from one company to the other (Mamnceko 8). Both the technical team and functional team apply data model. Technical team consists of programmers and developers while the functional team consists of the end user and business analysts. The data model is designed by individuals capable of meeting the expectation and providing a requirement for the team. Data modeling mostly uses the entity-relationship (E-R) diagramming format. In data modeling, the data element is numeric, a product can only be in one line at a time and region name of the customer is limited to a particular set of values. These facts are essential information in ensuring that data integrity internal database in the information system (Muller 119).

Data model components

Data model has inputs derived from the planning and analysis stage. At this stage, the data modeler collects the relevant information about the database requirements by interviewing the end users and reviewing the existing documents. Data model comprises of two outputs. The first output is the data document that describes the data objects, their relationships and the rules that are required by the database. The second output is the entity-relationship diagram that presents the structure of the data in a pictorial form.

Origin of data modeling

The key driver for the development of data modeling was the elimination of slow manual processes in organizations. Before the development of data modeling, the focus of system development focused on automating business processes that were done manually (Navathe 113). The development of technology led to complex changes in the business environment that in turn increases the complexity of the information system. Technological development has hindered the ability of the old method of conducting information system analysis to satisfy current business requirements. Data modeling can because of its ability to specify data structures in actual file followed with data management system such as databases. This led to the introduction of hierarchical and network models exemplified by data management system known as an integrated data store of network model and information management system of the hierarchical model. Later, the relational model was proposed as a mathematical model for data analysis and modeling. It provided a solution to a variety of problems related to databases. Moreover, it provided a solution for redundancy and a way for estimating the quality of the database structures in a formal way. The model made it possible for organizations to deal with distribution of data, data replication, security schemes and integrity constraints. Currently, the process of system analysis consists of two main activities namely functional modeling and data modeling. The object oriented methodologies emphasize data modeling by the use of class diagrams whereas traditional developmental methods focus on the functional modeling by applying data flow diagrams (Navathe 122).

Role of data modeling in system analysis

Data model is a blueprint for data entities implementation such as a relational database. It facilitates communication of the stakeholder during the analysis of the domain. A well-designed data model helps and organization to achieve performance requirements in a software system. In the information system, the data model is a crucial input to modifiability analysis. The data model should be created with an understanding of future extensions, increment developmental plans and data integration across information system. Data is a valuable asset in all organizations, and the existence of data model helps and organization to enforce a data quality needed for system analysis. The data modeling tools can generate application code required for accessing data tables, message schemas classes to hold the data and simple reports (Calì, Gottlob and Pieris 1889). The major goal of data modeling is to ensure that the data objects vital for the databases are represented accurately and completely. The data model can be easily reviewed and verified by the end users because it uses natural language and simple notations. Data model should include all the data required for the database and the procedures used should be correct and consistent. Moreover, data model should have the ability to accommodate changes as per the user requirements. Designing the right data model ensures that its application meets the need of the end user. Data model in system analysis presents many benefits in businesses such as reducing risks, reducing cost, provision of compatible data, increases effectiveness, and present business opportunities. Moreover, the data model is responsive to change and integrates business system Wand, (Wand, Storey and Weber 494). Data model consists of three phase’s namely structural, manipulating part and integrity part. The different part consists of a set of rules. The manipulating part consists of the types of the operations allowed including retrieving, updating and changing the database. The integrity part of the data model is used to validate data accuracy.

Types of data model

There is four basic data model at the various information levels namely conceptual, logical data model, enterprise, and physical data model. The conceptual data model provides a representation of an organization data. It served as a communication tool and considered as a model for identifying business entities or concepts and the relationship between these entities (Applied Information Science). The understanding of the relationship between business objects helps and organization to reflect, gain and document an understanding of the company based on its data. Data modeling in most businesses forms a crucial stage in the process of database design. Its purpose is to show the meaning of the data and their interrelationship. This type of data model is done together with and structuring steps and requirement analysis during the process of system analysis. It is carried out through the process of system development (Inforadvisors). The process is useful in not only planning but also the analysis phase of the life cycle in the system development. Conceptual data model process comprises of five steps namely design, implementation, maintenance, planning, and analysis. The model contains 10-20 entities and a relationship called group entities. The process starts with the development of a conceptual data model for the system that the organization wants to replace it already exists. Development of data is essential for the conversion of the existing database or current files into new system databases. Furthermore, it is a significant step for the determination of the data requirement for the new system (Chan 59; Tryfona and Jensen 245).

Entity-relationship modeling (E-R)

Most of the organizations perform conceptual data modeling by use of E-R modeling. It is a fundamental tool for the design of the database. The rapidly increasing interest in the object-oriented methods the use of class diagrams has also become popular. The E-R modeling notation uses three constructs namely data entities, relationships and their associated attributes (Chen 7). An entity refers to person, object, place, concept or an event in the user environment about which the company plans to maintain the data. It has been extended to include additional constructs. E-R modeling is a logical representation of a business area and is expressed regarding business environment entities, attributes of the entities, and their relationships. It is represented in an ER diagram and a graphical representation of the model. An ER diagram is a high-level logical model that is used by both the database designer and end user to document the data requirement by a business. It is a high-level model because it does not detailed information about and organizational data. It provides a conceptual understanding of the information (Dixit and Kumar 189; Burrow 6). The logical data model is applied in examining the concept of the domain and their relationship. It depicts the types of the logical entity referred to as entity types attribute describing data and entities relationship. It is an evolution of the conceptual data models towards the development of data management technology particularly the relational databases (Young-Gul & March, 103). The implementation of the one conceptual data model sometimes requires the application of multiple logical models (Burrow 6). There are two categories of the logical data model. They include physical data model and metadata models that are further divided into different models. The physical model is divided into business/application data models. The metadata model is divided into business intelligence metadata models, integration metadata models and unstructured metadata models (Applied Information Science). An enterprise data model is also referred to as external data model. It refers to an integrated view of data that is produced and used in the entire organization. It is unbiased of any application or system and represents an integrated definition of data. It does not depend on how organizational data is sourced, processed, stored or accessed. The structure of the enterprises data model comprises of three levels namely conceptual entity model, conceptual model, and subject area model. Enterprise data model depicts the data management system from individual user viewpoint. It is an appropriate model in holding data requirement for a particular organization and the rules that apply to it (Dey, Storey and Barron 453). Physical data model deal with how data are structured and maintained. It documents the technical details for implementation as either a data structure or database (Young-Gul and March 105). It is used to design the database internal schema depicting column name the data column tables, foreign key, primary key and their relationship. It takes into account the constraints and facilities of database management system. The physical data model provides various options such as message format, file structure, and physical schema. It indicates how an organizational data is stored physically, and it must support the conceptual model (Inforadvisors).

Data design in an organization

The design of data for an organization can be achieved through two approaches namely top down and bottom up approach. Top-down approach starts with a business need perspective by beginning with a very high-level design. The high-level data model is built by studying the existing business systems that may include operational systems or reporting systems. Conversely, the bottom-up approach does not take into account the need for the business. Instead, it focuses on the business systems environment. Thus, approach is more common in many businesses. However, the approach is ideal if there is ample system documentation and resources available are minimal. Moreover, it can be applied when the purpose of the data model is to gain a better understanding of the existing application (Wand, Storey and Weber 495). System analysis is a process of reconciling the needs of a business based on the information available. Therefore, many organizations are applying a hybrid approach that completes the initial step of gathering information by first applying top-down approach and then bottom-up analysis. This approach is the best when the business planning a new system or upgrading and existing system and require a business expert to carry out the process. A high-level data model is required in system analysis because it conveys organizations’ core principles and concepts in a simple manner. It uses concise description to and its simplicity makes it easier to understand for both technical and nontechnical individuals. Therefore, the purpose of a high-level data model in system analysis is to describe complex organization information. It presents benefits to an organization by creating a common goal, common understanding and the context of the core business concepts across functional business areas. This is critical for system analysis and success of management of data. It plays an essential role in the analysis of information systems that manages organizational data (Yan and Ma 1881).

Agent-based model

An agent-based model is also referred to as an individual-based model. They play a crucial role in system analysis through simulation of interaction and actions of autonomous agents that comprises both collective and individual entities such as groups or organizations. The main aim of simulating the entities is to investigate how they affect the operation of the system. The agent-based model combines the broad range of elements such as complex systems, multi-agent systems, game theory, emergence and evolutionary programming (Macal and North). Various techniques can be applied in designing an agent-based model. They include accessibility, modularity, management of complexity, refinability, openness, expressiveness and preciseness. The agent-based model has been applied as a modeling method in the system analysis to examine the relationship between level agent behavior and system level (Anylogic).

Agile and data modeling

Agile plays a crucial role in data modeling. It provides a solution to the traditional method of software development and management of a project (Hasnain and Hall 110). The existing practices for the development of software present some challenges. One of best practices that have been used is the waterfall approach that lack effective communication means. It has been widely used for the development of complex and critical software, but it is not flexible. The drawbacks of this methodology have led to the development of other approaches for data modeling such as agile. Agile methodology is appropriate for data modeling because it has the capacity to manage changes due to changes in simulation requirements. Agile data modeling method comprises of various techniques for function and data modeling that can be used in various ways (Sundararajan, Bhasi and Vijayaraghavan 246).

Data modeling development cycle

Data modeling is conducted in the early stages of a system development. Database lifecycle encompasses all the events that are undertaken from the time an organization recognize it needs a new database through its creation, deployment and the day the database is removed from the system. The system development life cycle comprises of several steps that include planning, gathering of the requirements, conceptual design, logical design, physical design, construction, implementation and ongoing support. Planning involves launching of a feasibility study that the project can achieve the objective of the business. Information that describes the project is gathered focusing on what the organization wants to achieve rather than how to achieve it. The conceptual design involves the development of the externals of the databases (Data Modeling Life Cycle). Technical design of the databases and applications to be included in the project are carried out during logical design. It is then converted to the actual system software and hardware needed for implementation of the databases and applications. In the construction phase, testing is carried out. The implementation phase involves installation of the components of the new system such as database objects, reports, and application programs. Following the implementation of the new system, ongoing support is required to identify the performance issues, failures and unexpected results. One of the biggest challenges encountered in developing a data model is ensuring that the project allows for proper use and development of high-quality data model. Another challenge is the development of a data model that provides sufficient information for making better decisions. A data model should be flexible to adapt to the new changes without having to change the code. Moreover, a data model should maintain and record all the data that is required for a system.

Importance of data model

The core characteristics that make data modeling valuable in system analysis are precision and communication. It is a representation of the organizational information requirements and thus it must reflect the data requirement of a given enterprise. It must clearly and precisely indicate every aspect of a business data. It serves as a communication tool for the domain experts or users. Therefore, the analysis and design of a data model must focus on communication and precision (Shoval, Danoch and Balabam 218). The most important factor in system analysis is quality. Currently, many organizations are facing challenges in maintaining their data. Data modeling has provided a solution for the system analysis by enabling the managers to view the organizations from all perspectives. Markets are becoming increasingly complicated, and for an organization to succeed and survive, it must make a better decision based on the business data. Data modeling has made it easier for organizations to make a decision by providing them with an opportunity to evaluate and enhance efficiency. Modeling the data has improved the quality of data in the system by improving the process for making decision (Chen 9).


Data modeling is a powerful tool for representing and documenting organizational data. It is a crucial aspect of system analysis that plays a key role in data warehousing. A data model provides guidelines for building an organizational database. However, it must be simple to create a common understanding and goal. Data modeling must depict the structure of the data system serving various purposes such as identification of some of basic aspects of the system regarding the entities and their role in the organization. The ER diagram model is a widely used in organizations to build data models. There are four types of data model and the type of the model to be used by a particular organization depends on the available resources and the purpose of the model.