Systems Analytics and Enterprise Management 4

Alternative Matrix

An  alternative matrix  can be used to organize the pros and cons of the design alternatives so that the best solution will be chosen in the end ( Figure 7‐6 ). This matrix is created by the same steps as the feasibility analysis, which was presented in  Chapter 1 . The only difference is that the alternative matrix combines several feasibility analyses into one matrix so that the alternatives can be easily compared. The alternative matrix is a grid that contains the technical, economical, and organizational feasibilities for each system candidate, pros and cons associated with adopting each solution, and other information that is helpful when making comparisons. Sometimes, weights are provided for different parts of the matrix to show when some criteria are more important to the final decision.

FIGURE 7-6  Sample alternative matrix using weights.

To create the alternative matrix, draw a grid with the alternatives across the top and different criteria (e.g., feasibilities, pros, cons, and other miscellaneous criteria) along the side. Next, fill in the grid with detailed descriptions about each alternative. This becomes a useful document for discussion because it clearly presents the alternatives being reviewed and comparable characteristics for each one.

Sometimes, weights and scores are added to the alternative matrix to create a  weighted alternative matrix  that communicates the project's most important criteria and the alternatives that best address them. A scorecard is built by adding a column labeled “weight” that includes a number depicting how much each criterion matters to the final decision. Typically, analysts take 100 points and spread them out across the criteria appropriately. If five criteria were used and all mattered equally, each criterion would receive a weight of 20. However, if cost were the most important criterion for choosing an alternative, it may receive 60 points, and the other four criteria may get only 10 points each.

Then, the analysts add to the matrix a column called “Score” that communicates how well each alternative meets the criteria. Usually, number ranges like 1–5 or 1–10 are used to rate the appropriateness of the alternatives by the criteria. So, for the cost criterion, the least expensive alternative may receive a 5 on a 1–5 scale, whereas a costly alternative would receive a 1. Weighted scores are computed with each criterion's weight multiplied by the score it was given for each alternative. Then, the weighted scores are totaled for each alternative. The highest weighted score achieves the best match for our criteria. When numbers are used in the alternative matrix, project teams can make decisions quantitatively and on the basis of hard numbers.

It should be pointed out, however, that the score assigned to the criteria for each alternative is nothing more than a subjective assignment. Consequently, it is entirely possible for an analyst to skew the analysis according to his or her own biases. In other words, the weighted alternative matrix can be made to support whichever alternative you prefer and yet retains the appearance of an objective, rational analysis. To avoid the problem of a biased analysis, each analyst on the team could develop ratings independently; then, the ratings could be compared, and discrepancies resolved in an open team discussion.

YOUR TURN 7‐2 Weighted Alternative Matrix

Pretend that you have been assigned the task of selecting a CASE tool for your class to use for a semester project. Using the Web or other reference resources, select three CASE tools (e.g., Visible Analyst Workbench, Oracle Designer). Create a weighted alternative matrix that can be used to compare the three software products in the way in which a selection decision can be made. Have a classmate select the “right” tools, according to the information in your matrix.

The final step, of course, is to decide which solution to design and implement. The approval committee should make the decision after the issues involved with the different alternatives are well understood. Remember that the line between the analysis and design is quite fuzzy. Sometimes alternatives are described and selected at the end of analysis, and sometimes this is done at the beginning of design. The bottom line is that at some point before moving into the heart of the design phase, the project team and the approval committee must understand all of the feasible ways in which the system can be created, and they must select the way that makes the most sense for the organization. The acquisition strategy selection that is made will then drive many of the remaining activities in the design phase.

Applying the Concepts at DrōnTeq

Jiang Tsiao, senior systems analyst and project manager for DrōnTeq's Client Services system, had three different approaches that he could take with the new system: he could develop the entire system, using development resources from DrōnTeq; he could buy a packaged software program (or a set of different packages and integrate them); or he could hire a consulting firm or service provider to create the system. Immediately, Jiang ruled out the third option. Building these Internet‐based applications was becoming increasingly important to the DrōnTeq business strategy. By outsourcing these Internet applications, DrōnTeq would not develop Internet application development skills and business skills within the organization.

Instead, Jiang decided that this system would be developed primarily using custom development with the company's standard Web development tools. In this way, the company would be developing critical technical and business skills in‐house, and the project team would be able to have a high level of flexibility and control over the final product. Jiang believed that the strategic importance of the new Client Services business unit and the critical role the information system he was overseeing justified the time and expense of a custom developed system. There were some components, however, that he felt could be obtained through purchased software products that would be integrated into the overall system.

As mentioned previously, there was at least one part of the project that might be handled by packaged software: the website that would be used to provide information to potential flight service clients and to prospective drone pilots. Jiang realized that a multitude of programs have been written and are available (at low prices) to handle development of content‐rich, informational websites. These programs, called website builders, usually allow users to build websites quickly, are generally easy to use, and provide an array of templates and optional add‐in components. Jiang believed that the project team should at least consider some of these packaged alternatives so that less time would be spent handling basic Web tasks and more time could be devoted to the unique, strategic elements of the client services system.

To help better understand some of the website builder programs that were available in the market and how their adoption could benefit the project, Jiang asked Maria, another team member, to research some options and narrow the list of candidates down to three. Maria found numerous options, but many were eliminated as not robust enough to meet DrōnTeq's standards. Maria consulted with Rahul, the team's infrastructure analyst, to help her create the evaluation criteria for this software tool. She then recruited Dawn, another team member, to work with her to review and assign scores to the final three candidate products.

Maria created a weighted alternative matrix that compared three different website builder programs against one another ( Figure 7‐7 ). She and Dawn determined the relative importance of the evaluation criteria to the project, and then assigned their scores and calculated each product's weighted average score. Although all three alternatives had positive points, Maria could see that alternative 3 (WB‐3) was the best alternative for handling this project. WB‐3 is based on PHP, a Web scripting language, and incorporates MySQL, an open‐source relational database; both are included in DrōnTeq's standard Web development environment. While the cost of WB‐3 was higher than the other products and it is difficult to learn initially, the high scores on all the technical evaluation criteria were persuasive. Maria made a note to look into acquiring WB‐3 as the Web development platform to build the informational component of the Client Services system.

Evaluation Criteria	Relative Importance (Weight)	Alt 1: WB‐1	Score (1‐5) *	Wtd Score	Alt 2: WB‐2	Score (1‐5) *	Wtd Score	Alt 3: WB‐3	Score (1‐5) *	Wtd Score
Technical Issues:
Integration with existing infrastructure	15	Very little capability	2	30	Provided, but appears awkward	3	45	Strong, appears seamless	5	75
Database capabilities	15	None	1	15	Limited	2	30	Excellent; compatible with company standards	5	75
Access to underlying code	10	Not possible	1	10	Limited	3	30	Easy	5	50
Video support	15	Yes; adequate	3	45	Yes; adequate	3	45	Yes; excellent	5	75
Economic Issues:
Cost	20	$15/month	5	100	$25/month	4	80	$90/month	1	20
Organizational Issues:
Market adoption	5	Strong ‐ widely used	4	20	Moderate ‐ newer product	3	15	Strong ‐ market leader	5	25
Ease of learning	10	High	5	50	Somewhat complex	3	30	High learning curve	1	10
Ease of use	10	Inflexible	2	20	Somewhat flexible	4	40	Very flexible; easy to modify	5	50
TOTAL	100			290			315			380

* The score denotes how well the alternative meets the criteria; 1 = poor fit; 5 = perfect fit.

FIGURE 7-7  Alternative matrix for website builder program.