Building Web-Based and Inter-Organizational Decision Support Systems

Problem solving and knowledge management go hand-in-hand. Together they have become one of the most important aspects of organizational decision making. Managers around the world realize that much of their organizations’ value depends on their ability to gather, analyze and manage knowledge and use it to solve problems.

To accelerate the process of transforming information into knowledge and applying it to resolve problems, organizations use resources in addition to human intellect. These resources help them discover patterns, identify context where these patterns work and analyze alternative solutions.

A computerized decision support system does all the work and aids managers in making the right choice. However, with the arrival of the internet and modern communication technologies, the organizational decision support systems deploy web-based technologies. The new science of gathering and distributing information adds efficiency to DSS, ultimately helping managers make more appropriate decisions.

The best part is that the web-based decision support systems can be knowledge-driven, communications-driven, model-driven, document-driven or data-driven or hybrid. Web technologies can be integrated with any type of decision support system. Since internet technologies are known for effective, faster and safer information distribution, the web-based DSS can be used across the organization and by two or more organizations where there is a need for information sharing.

Key Terms

Before we move on and get into the intricacies involved in building a web-based decision support system, it’s important to understand these terms:

• W3 or www: It stands for the World Wide Web, which in itself is an information system of interlinked hypertext documents that are accessed via the internet. Commonly, hypertext documents are called web pages. Generally W3 and internet are used interchangeably. But they are not same. Internet is a system of interconnected computer networks whereas W3 is one of the services for these networks. The latter is a collection of images, text documents and other resources that are linked by URLs and hyperlinks.

• Web-based DSS: Web-based DSS is a computerized system to deliver information for decision support using a ‘thin-client’ browser like Internet Explorer or Netscape Navigator. The computer server that hosts DSS application is linked to the user’s computer by a network using TCP/IP (Transmission Control Protocol/Internet Protocol).

• Inter-Organizational DSS: An inter-organizational DSS is a system that’s used by organizations working together on a particular problem or project. A DSS used by a company’s investors, bankers, stockholders, suppliers and customers is an example of an inter-organizational DSS. Web technologies drive communication and information distribution.

• Intranet: It is a private network that can be accessed only by an organization’s staff. A DSS used across the organization facilitates information sharing through intranet.

• Extranet: It’s not a private network but allows controlled access to information to vendors, partners, suppliers and customers.

• TCP (Transmission Control Protocol): It’s the core protocol of the IPS (Internet Protocol Suite), to provide trustworthy, controlled, prearranged delivery of digital information units between applications running on hosts, over an IP network. TCP is a protocol on which emailing, W3, FT (file transfer) and remote administration applications rely on.

Designing and Developing Web-Based DSS

Most companies simply integrate a web browser on their existing decision support systems, in the name of upgrading it and deploying a so-called web-based DSS. The results, most likely, are unsatisfactory and ambiguous. This is because the web technologies are integrated without carrying a feasibility analysis. When a DSS is initially designed, it focuses on the current requirements of its users. Ideally, managers shouldn’t expect it to be efficient in other areas.

So, the question remains - what goes into designing and developing a web-based decision support system?

Here is the step-by-step process explaining how a web-based system is designed and developed. However, this is a standard process and any of the below mentioned steps can be removed during system development, depending on the understanding and preferences of developers, analysts and users.

1. Problem Identification: This is universal because without having to know what you want a web-based DSS to do for you, it’s impossible to move ahead – let alone getting things right. Brainstorm with your team members to identify the problems you want to resolve using a DSS.

2. Conceptualization: Next step is to conceptualize the idea on which your DSS will be based. Ask yourself these questions:

   • What is it going to do?

   • How is it doing to do?

   • What techniques will it use?

   • Who’s going to use it?

   • What processes do you want it to carry?

3. Feasibility Analysis: Once you have a clear idea of a prospective decision support system, you must assess it. Try to uncover

   • Its strengths and weaknesses

   • Opportunities and threats

   • Resources required to build it

   • Prospects for success

   In addition to this, a feasibility analysis sheds light on a project’s operational, economical, technological, scheduling and legal viability.

4. System Development: Once the prospective system has passed your feasibility test, you can begin with system development. Remember that system development doesn’t just require technical expertise; rather it’s a collaborative effort of managers, DSS analysts, developers and finance professionals. The focus must be on:

5. UI Development: Developing a user interface remains the most crucial aspect of any computer-based system. And it becomes more important when a prospective system will be used by individuals of various levels of sophistication and technical know-how. Control, memory aids and suggestions based on user’s history are important components of a user interface, when it comes to a web-based system.

6. DSS Architecture: User interface is just a part of DSS architecture and thus, is not separated from it. However, we mentioned it separately because UI is like a middleman that fosters communication between the system and the user. Arguably, it’s the most important aspect of DSS architecture.

   Typically a web-based DSS is built on three or four tier architecture. One, user sends a request through a web browser using HTTP to a web server. Two, the web server then uses a program or a script to process the request. Three, the script may link to a model, processes a database request or format a document. Four, the web browser where the user sent a request displays results most suited to his/her query.

7. Tool Selection: HTML (Hyper Text Markup Language) is the most commonly used tool for developing web-based DSS. HTML is not a programming language, but can be used for input and output from a decision support which is programmed in Java or JavaScript. HTML has been basically designed to stipulate the rational organization of documents with hypertext extensions.

   XML (Extensible Markup Language), CGI (Common Gateway Interface Scripts), JavaScript code in HTML pages, Java applets and ActiveX are other tools that can be used to develop web-based DSS. In order to weigh your options, you must consult your developer and determine development costs.

8. System Implementation: Once the system is developed, the next phase is to implement it. A number of tools are available that specialize in implementation of DSS. These are ColdFusion and dbProbe.

Potential Problems

When an organization gets on with web-based DSS, there can arise many potential problems, which can be easily resolved. Let’s take a look at potential problems and their solutions.

• As the DSS is web-based, it may encounter peak load problems when many managers use the system simultaneously. High performance hardware can resolve this problem.

• Web is stateless and doesn’t keep track of configuration settings automatically. At the same time users don’t want to reenter the username and password each time they use the system. You must consult your developers or solution providers on how to handle user authorization and authentication.

• Users must learn to use the system rapidly. This is because web technologies keep advancing and there will be a need to upgrade the system on regular basis. Training can help users keep up with the changing technology.

Managing Web-Based and Inter-Organizational Decision Support System

Managing web-based and inter-organizational decision support systems can be daunting. This is because these systems are used by a many individuals for making shared decisions. Despite observing extreme caution when developing a system, there are some real-world issues that are bound to arise. Reengineering business process can be really challenging. It’s not an easy task to redesign or reengineer a business process when it involves a huge number of users. Managers in interdependent organizations must consider following issues to ensure effective implementation of a web-based DSS:

• The first major issue is – who will use the system: managers, suppliers, customers or all? When you answer this question, you’ll identify the associated questions: whether there will be a need to redefine processes; how are you going to train the users to work on a DSS; what are the chances of its success, etc. If the answers to these questions are ambiguous, your DSS is destined to fail.

• The second major issue is availability of technical talent to develop a web-based DSS. Experts need to determine hardware, software and manpower required to build a system. So, managers need to ask themselves if they have in-house capabilities to work on developing the system. If no, where are they going to find their technology partners?

• Third major issue is legality of the information distributed through DSS to external parties. Consider, if you are likely to face any copyright or privacy issues?

Examples of Web-Based DSS

There are several famous web-based decision support systems, including but not limited to:

• Microsoft Carpoint: Demonstrating both data and model-driven DSS, Microsoft Carpoint is a web-based system that allows users to make pair-wise comparisons of car models.

• TCB Works: A communication-driven web-based DSS, TCB Works enables people to interact, discuss and make both structured and multi-criteria decisions. Users are prompted to enter username and password, in order to start with the project screen.

• Fidelity ‘Retirement Planning Calculator’: It is a model-driven DSS to help a person decide how much he or she needs to invest each month for retirement.

• Netscape Decision Guides: These are model and knowledge-driven web-based decision support systems, offering 25 different decision guides on diverse topics, including choosing pets, bikes and business schools.

• Stockfinder: It’s a data-driven DSS to help investors identify stocks based on various criteria, including industry type, price and earnings.

Companies with Web-based DSS

Many companies across the world have implemented some kind of web-based decision support systems, including but not limited to:

• NDC Health Information Services has created a web-enabled prescription data warehouse that it sells to pharmaceutical manufacturers.

• Bell Canada uses the Essbase Web Gateway enabling hundreds of sales and business managers compose their own interactive queries from their own web browsers. They can also navigate, analyze and update sales forecasts for their own use.

• Societe Generale USA installed a multi-tier architecture that enables the support of both web computing and client server.

• Hannaford brothers Grocery chain uses a DSS that provides store managers with detailed reports sales, cost, inventory and budget, so that they can use the information to make decisions.

Advantages and Disadvantages of Web-based DSS

Web-based decision support systems though are the latest in the class. But they also have a fair share of pros and cons. Let’s take a look at their advantages and disadvantages:

Advantages

• Web-based DSS reduce decision making costs to a great extent.

• They have reduced geographical and technological barriers.

• They make it easy for companies to involve their vendors, suppliers and customers in decision making, surveys, etc.

• They have significantly improved the speed of information distribution.

• They provide an excellent way to create and manage a knowledge repository.

• They have reduced end-user training costs as web technologies nowadays are used by almost everyone.

Disadvantages

• Web-based DSS are extremely efficient, due to which users sometimes set unrealistic expectations.

• They may succumb to peak demands and experience load problems.

• Web-based DSS require additional security, which may be expensive.

• Web technologies advance at a very fast pace. It may be difficult to upgrade the system so frequently.

• The knowledge repository may accumulate obsolete reports, if users don’t delete old information.

The internet and World Wide Web have created major opportunities to create, manage and share both quantitative and qualitative information while keeping the costs low. Though these technologies don’t resolve all the problems but they have contributed significantly to knowledge management and decision making.

Web-based DSS can be very effective if managers are aware of what they want to create and how they want to use it. These systems are expected to evolve and become smarter with the advancement in internet technologies.