Methods Available

PREPARING THE SYSTEMS PROPOSAL METHODS AVAILABLE The systems proposal is a distillation of all that the systems analyst has learned about the business and what is needed to improve its performance. In order to address information requirements adequately, the systems analyst must use systematic methods for acquiring hardware and software; identify and forecast future costs and benefits and perform cost-benefit analysis, as shown in Figure 12.1. All of these methods are used in preparing systems proposal material. Information needs of users drive the selection of computer hard- wale, data storage media, and any prepackaged software. The hardware and software system that is eventually proposed is the analyst’s response to users’ information needs. This chapter provides the methods necessary to project future needs systematically and then weigh current hardware and software alternatives. Forecasting, guidelines for hardware and software acquisition, and cost-benefit analysis are included.

ASCERTAINING HARDWARE AND SOFTWARE NEEDS In this section, we cover the process of estimating the present and future workloads of a business, and the process involved in evaluating computer hardware and software to handle workloads adequately. Figure

FIGURE 12.1 Major activities In preparing the systems proposal

12.2 shows the steps the systems analyst takes in ascertaining hardware and software needs. First, all current computer hardware must be inventoried to discover what is on hand and what is usable. Then current and future system workloads must be estimated. Following this, an evaluation of available hardware and evaluation of software is undertaken. The systems analyst needs to work along with users to determine what hardware will be necessary. Hardware determinations can come only in conjunction with determining information requirements. Knowledge of the organizational structure (as discussed in Chapter 2) can also be helpful in hardware decisions. Once systems analysts, users, and management have a good grasp of what kinds of tasks must be accomplished, hardware options can be considered.

Inventorying Computer Hardware Begin by inventorying what computer hardware is already available in the organization. As will become apparent, some of the hardware options involve expanding or recycling current hardware, so it is important to know what is on hand. If an updated computer hardware inventory is unavailable, the systems analyst needs to set up one quickly and carry through on it. You need to know:

FiGURE 12.2 Steps In choosing hardware and software 1. Type of equipment—model number, manufacturer 2. Status of equipment operation—on order, operating, in storage, in need of repair 3. Estimated age of equipment 4. Projected life of equipment 5. Physical location of equipment 6. Department or person considered responsible for equipment 7. Financial arrangement for equipment-—owned, leased, or rented

Figure 12.3 provides an example of a hardware inventory form that can be given to the relevant people. It should be easy to fill out, and 1. Equipment a. CPU/type manufacturer/model b. Storage/type/manufacture/model C. Input/type manufacturer/model d. Output/type/ manufacturer/model e. Communications type/manufacturer/model 2. Status On order

Fully operational In need of repair In storage 3. Estimated Age of Equipment 4. Projected Life 5. Located 6. Name of individual responsible for equipment: Purchase Operation Maintenance 7. Financial Agreement Owned Rented Leased FIGURE 12.3 Using a form for taking inventory of equipment currentiy available In the organization

should explicitly request that peripherals such as disk drives and VDTs be enumerated separately. Ascertaining the current hardware available will aid in a sounder decision-making process when hardware decisions are finally made, since much of the guesswork about what exists will be eliminated. Through your earlier interviewing, questionnaires, and research of archival data, you will already know the number of people available for data processing as well as their skills and capabilities. Use this information to project how well the staffing needs for new hardware can be met.

Estimating Workloads The next step in ascertaining hardware needs is to estimate workload. This means that systems analysts can formulate a number that represents both current and projected workload for the system, so that any hardware obtained will possess the capability to handle current and future workloads.

If estimates are accomplished properly, the business should not have to replace hardware solely due to unforeseen growth in system use. (However, other events, such as superior technological innovations. May dictate hardware replacement if the business wants to maintain its competitive edge). Out of necessity, workloads are sampled rather than putting actual workloads through several computer systems. The guidelines given on sampling in Chapter 4 can be of use here, since in workload sampling the systems analyst is taking a sample of necessary tasks and computer resources required to complete them. Figure 12.4 is a comparison of the workload between an existing and proposed information system. Notice that the company currently is using a manual system to make a monthly summary of shipments to their distribution warehouses, and a computer system is being suggested. The workload comparison also looks at cost per hour of each system; when and how each process is done; how much human time is required; and how much computer time is needed.

Evaluating Computer Hardware Evaluating computer hardware is the shared responsibility of management, users, and systems analysts. Although vendors will be supplying detail about their particular offerings, analysts need to oversee the evaluation process personally since they will have the best interests of the business at heart. Additionally, systems analysts may have to educate users and management about general advantages and disadvantages of hardware before they can capably evaluate it. Based on current inventory of computer equipment and adequate estimates of current and forecast workload, the next step in the process is to consider kinds of equipment available that appear to meet projected needs. Information from vendors on possible systems and system configurations becomes more pertinent at this stage, and should be reviewed with management and users. Additionally, workloads can be simulated and run on different systems, including those already used in the organization. This is referred to as benchmarking. Criteria that the systems analysts and users should use to evaluate performance of different systems hardware include: time required for average transactions (including how long it takes to input data and how FIGURE 12.4 Comparisons of workloads between, existing and proposed systems

long it takes to receive output); total volume capacity of the system (how much can be processed at the same time before a problem arises); idle time of the central

processing unit; and size of memory provided. Some criteria will be shown in formal demonstrations; some cannot be simulated and must be gleaned from manufacturer is specifications. It is important to be clear about required and desired functions before getting too wrapped up in vendors claims during demonstrations. Once functional requirements are known and the current products available are comprehended and compared with what already exists in the organization, decisions are made by the systems analysts in conjunction with users and management about whether obtaining new hardware is necessary. Options can be thought of as existing on a continuum from using only equipment already available in the business all the way to obtaining entirely new equipment. In between are options to make minor or major modifications to the existing computer system.

Computer Size And Use The rapid advance of technology dictates that the systems analyst research types of computer available at the particular time that the systems proposal is being written. Therefore, we simply provide an overview of available sizes of computers and their typical uses rather than dwell on features of specific computer models. Computer sizes range all the way from the smallest lap-sized microcomputers to room-sized supercomputers. Each has different attributes to consider when deciding how to implement a computer system. Microcomputers. Microcomputers seem ubiquitous, and with their increased portability they can indeed be used almost anywhere. Acceptability of microcomputers as a tool for the individual to own was furthered by dubbing them “personal computers.” Microcomputer systems typically consist of a keyboard or mouse as a user interface: a VDT for input; a simple operating system: limited primary storage; the same VDT and a printer for output secondary storage on floppy diskettes; and/or a hard disk. Microcomputers are useful for individuals, for small businesses with limited data storage needs, or for automating small systems such as home security. Minicomputers. A minicomputer is a little larger than the micro. joining minicomputers together through networking can result in computer power rivaling that found in a large scale computer, and may actually super scale a large scale computer in the flexibility it affords. Minicomputers are more expensive than microcomputers, but they offer superior networking and storage capabilities. Some businesses also view minicomputers as more business oriented than microcomputers, but this opinion is gradually eroding as more business application software becomes available for microcomputers.

Medium scale computers. Medium scale computers are one step above minicomputers in size and offer concurrently more capacity and speed. Rather than needing to network a medium scale computer, a business would use its capabilities to process many jobs at the same time. Large scale computers. Large scale computers have faster capabilities than the medium scale ones. They process numerous jobs from many different sites simultaneously. Their storage capacity is huge. Supercomputers. The last size of computer we consider is the super computer. Due to the size and cost of supercomputers, few analysts have as yet had the experience of working on projects involving them. Super computers are typically shared by different user groups for different applications. For example, a group of university geologists and a group of companies from the oil industry might band together to purchase a supercomputer, taking advantage of its incredibly fast calculating ability and immense primary storage to work on joint interests.

Alternative Date Storage Devices Storage of data, either before or after processing, is extremely important to the smooth functioning and eventual cost of the system. Generally, the choice of storage medium is determined by the kinds of processing needed. Key questions include how long data will be kept until processing, how large files are, and what kind of access to stored data is necessary. Magnetic tape. Magnetic tape is one of the older types of computer storage and is a commonly used one. Its main advantage is relatively low cost per record stored. So that it is a feasible storage medium for high volume applications. Its chief disadvantage is the inability to access one record quickly for updating or query, since data are stored and read in blocks, as shown in Figure 12.5. In order to update a single record, the entire tape would

FIGURE 12.5 How data are accessed from magnetic tape

have to be reread and rewritten. Magnetic tape is useful for applications in which data are saved to be run later. Disks. Disks are also a magnetic medium, but they are better suited to storage of information that must be received quickly. In Figure 12.6, you can see that disks are arranged in sectors and tracks. This emphasizes one of the main advantages of disks over magnetic tape their use of sector/track organization for accessing a few records quickly, rather than having to read through an entire tape sequentially. This

means that disks are better suited than magnetic tape to low-volume applications. Additionally, disks are preferable for real-time uses, since records can be a altered quickly and new data can then be used to carry out further transactions correctly. Floppy diskettes. Floppy diskettes (or floppies) have similar characteristics to disks. Floppies are also referred to as flexible diskettes because the 5 ¼ size bends. Diskettes that are 3 ½ are encased in plastic, but are the same floppy material inside. Floppies are used with micro

FIGURE 12.6 How data are accessed from disks

computers and are good for low-volume uses. Their chief advantage is portability and low cost. Their main disadvantages are Low storage Capacity per floppy (which necessitates constantly changing diskettes and makes backing up tedious), and the ability to process only one file at a time. Hard disk drives. The only other reasonable storage and processing alternative for small computer systems is use of a hard disk, which has sealed within it both storage capacity and reading capacity. The chief advantages of hard disk drives is their greatly expanded storage capacity and their ease of handling compared to floppy diskettes. Hard disks can be either fixed or removable. Their chief disadvantages are their relatively high upfront price and the ultimate necessity of somehow backing up hard disk files.

Acquisition Of Computer Equipment The three main options for acquisition of computer hardware include buying, leasing, or renting it. There are advantages and disadvantages to weigh for each of the decisions, as shown in Figure 12.7. Some of the

FIGURE 12.7 Comparing the advantages and disadvantages of buying, leasing, and renting computer equipment

more influential factors to consider in deciding which option is best for a particular installation include: initial versus long-term costs; whether the business can afford to tie up capital in computer equipment; and whether the business desires full control of and responsibility for the computer equipment.

Buying implies that the business will itself own the equipment. One of the main determinants of whether to buy is the projected life of the system. If the system will be used longer than four to five years (with all other factors held constant). The decision is usually made to buy. Notice in the example in Figure 12.8 that the cost of purchase after six years is dramatically lower than that of leasing or renting. As systems become smaller and distributed systems become increasingly popular, more businesses are deciding to purchase equipment. Leasing, rather than buying, computer hardware is another possibility. Leasing equipment from the vendor or a third party leasing company is more practical when the projected life of the system is less than four years. Additionally, if significant change in technology is imminent, leasing is a better choice. Leasing also allows the business to put its money elsewhere, where it can be working for them, rather than being tied up in capital. Over a long period, however, leasing is not an economical way to acquire computer power. Rental of computer hardware is the third main option for computer acquisition. One of the main advantages of renting is that none of the company’s capital is tied up. Hence, no financing is required. Also, renting computer hardware makes it easier to change system hardware. Finally, maintenance and insurance are usually included in rental agreements. However, because of the high costs involved, and the fact that the company will not own the rented equipment, rental should be contemplated only as a short-term move to handle nonrecurring or limited computer needs, or when technological changes are volatile.

Evaluation Of Vendor Support For Computer Hardware There are several key areas to evaluate when weighing the support services available to businesses from vendors. Most vendors offer testing of

FIGURE 12.8 Comparison of alternatives for computer acquisition

hardware upon delivery and a 90-day warranty covering any factory defects, but you must ascertain what else the vendor has to offer. Vendors with comparable quality frequently distinguish themselves from others by the range of support services they offer. A list of key criteria to check when evaluating vendor support is provided in Figure 12.9. Most of the extra vendor support services listed there are negotiated separately from hardware lease or buy contracts.

Support services include routine and preventive maintenance of hardware; specified response time (within six hours; next working day, etc.) in case of emergency equipment breakdowns; loan of equipment in the event that hardware must be permanently replaced or off-site repair is required; and in-house training, or group seminars off site for users. Remember that it may be more difficult to obtain training for unique hardware that is not widely used by other organizations. While the possibility of a customized installation may be attractive, the prospects for its long term support maybe diminished. Peruse the support services accompanying the purchase or lease of equipment, and remember to involve appropriate legal staff before signing contracts for equipment or services. Unfortunately, evaluating computer hardware is not as straightforward as simply comparing costs and choosing the least expensive option. Some other eventualities commonly brought up by users and management include: (11 the possibility of adding onto the system if the need comes up later; (2) the possibility of interfacing with equipment from

FIGURE 12.9 Guidelines for vendor selection Vendor Selection Criteria 1. Hardware Support Full line of hardware Quality products Warranty 2. Software Support Complete software needs Custom programming Warranty 3. Installation and Training Commitment to schedule In house training Costumer hot line 4. Maintenance

Routine maintenance procedures Specified response time in emergencies Equipment loan while repair is being done

other vendors if the system needs to grow; (3) the benefits of buying more memory than is projected as necessary, with the expectation that the business will eventually “grow into it”; and (4) corporate stability of the vendor. Adding on to the existing system often is the spur for systems projects. Installing a system with add-on capability is a worthwhile way to proceed. Although it takes a little extra planning, it is cheaper and more flexible than the third approach of obtaining excess memory and carrying it in inventory for a number of years. Competition among vendors has made the idea of producing hardware that is compatible with a competitor’s important for vendors’ survival. However, before becoming convinced that buying cheaper compatibles is the way to endow your system with add-on capability, do enough research to feel confident that the original vendor is a stable corporate entity.

Software Evaluation Packaged software, rather than application programs specifically written for an installation, are becoming more readily available and certainly should be given careful consideration. Many hours of valuable programmer time can be saved if packaged software is deemed suitable for part or all of the system and extensive customizing isn’t necessary. The bar chart in Figure 12.10 shows the cost of software (projected as ever-increasing) as part of total hardware/software costs.

FIGURE 12.10 The cost of software as percentage of total hardware/software costs is rising

Software Evaluation 1. Performance Effectiveness Able to perform all required tasks

Able to perform all tasks mat may be desired at some time in the future Wall-designed VDT screens Adequate capacity 2. Performance Efficiency Fast response time Efficient input Efficient output Efficient storage of data Efficient back-up 3. Ease of Use Satisfactory user interface Help menus available Flexible interface Adequate feedback Good error recovery 4. Flexibility Options for input Options for output Usable with other software 5. Quality of Documentation Good organization Adequate tutorial Answers questions adequately 6. Manufacturer Support Hot line Newsletter

Frequent (low-cost) updates FIGURE 12.11 Guidelines for evaluating software

Once again, you will b dealing with vendors who may have their own best interests at heart. You must be willing to evaluate software along with users and not be unduly influenced by vendors’ sales pitches. Specifically, there are six main categories on which to grade software, as shown in Figure 12.11: performance effectiveness and efficiency, ease of use, flexibility, quality of documentation, and manufacturer support. Evaluate packaged software based upon a demonstration with test data from the business considering it and an examination of accompanying documentation. Vendors’ descriptions alone will not suffice. Vendors typically certify that software is working when it leaves their supply house, but they will not guarantee that it will be error-free in every instance or will not “crash” when incorrect actions are taken by users. Obviously, they will not guarantee their packaged software if used in conjunction with faulty hardware. The need for multiple copies of software (for use at several microcomputer workstations, for instance) means negotiating a multiple use agreement with the vendor so that copyrights are not infringed through

Consulting Opportunity 12.1 Veni, Vidi, Vendi or I Came, I Saw, I Sold “It’s really some choice. I mean, no single package seems to have everything we want. Some of them come darn close, though,” says Roman, an advertising executive for Empire Magazine, with whom you have been working on a systems project. Recently, the two of you have decided that packaged software would probably suit the advertising department is needs and stem its general decline. “The last guy’s demo we saw, you know, the one who worked for ‘Data Coliseum,’ really had a well-rounded pitch. And I like their brochure. Full color printing, on card stock. Classic,” Roman asserts.

“And what about those people from ‘Vesta Systems?’ They are really fired up. And their package was easy to use with a minimum of ceremony. Besides, they said they would train all twelve of us, on site, no charge. But look at their advertising. They just take things off their printers. It is not even letter quality.” Roman fiddles in his chair as he continues his ad-hoc review of software and software vendors. “That one package from Mars, Inc., really, sold me all on its own, though. I mean, it had a built-in calendar. And I like the way the menus for the screen displays could all be chosen by Roman numerals, It was easy to follow. And the vendor isn’t going to be hard to move on price. I think they’re already in a price war. “But, do you want to know my favorite?” Roman asks archly. “It’s the one put out by Jupiter, Unlimited, I mean, it has everything, doesn’t it? It costs a little extra coin, but it does what we need it to do, and the documentation is heavenly. They don’t do any training, of course they think they’re above it.” You are already plotting that in order to answer Roman’s burning questions by your March 15 deadline, you need to evaluate the software systematically as well as the vendors, and then render a decision, Evaluate each vendor/package on what Roman has said so tar (assume you can trust his opinions). What are Roman’s apparent biases when evaluating software and vendors? What further information do you peed about each company and its software before making a selection?

the creation of illegal copies. This often means purchase of one software package at its regular price and purchase of any additional copies at a reduced price. It is also possible to negotiate a special vendor services contract covering support for purchased software. This might include emergency and preventive maintenance; free or reduced-price updates; additional copies of documentation; and special user training.

IDENTIFYING AND FORECASTING COSTS AND BENEFITS Costs and benefits of the proposed computer system must always be considered together, since they are interrelated and often interdependent. Although the systems analyst is trying to propose a system that fulfills various information requirements, decisions to continue with the proposed system will be based on a costs and benefits analysis, not on information requirements. In mans’ ways, benefits are measured by costs, as will become apparent in the next section.

Forecasting Costs And Benefits Systems analysts are required to predict certain key variables before the proposal is submitted to the client. To some degree, a systems analyst will rely on “what-if” analysis: for example, “What if labor costs rise only 5 percent per year for the next three years, rather than 10 percent?” The systems analyst should realize that you cannot rely on “what-if” analysis for everything if the proposal is to be credible, meaningful, and valuable. The systems analyst has many forecasting models available. Figure 12.12 represents a decision tree for choosing one of the forecasting methods. The main condition is the availability of historical data. If they are unavailable, the analyst must turn to one of the judgment methods including estimates from the sales force, surveys to estimate customer demand, Delphi studies (a consensus forecast developed independently by a group of experts through a series of iterations), creating scenarios, or drawing historical analogies. If historical data are available, the next differentiation between classes of techniques involves whether the forecast is conditional or unconditional. Conditional implies that there is an association among variables in the model or that a causal relationship exists. Common methods in this group include correlation, regression, leading indicators, econometrics, and input-output models. Unconditional forecasting means the analyst isn’t required to find or identify any causal relationships. Consequently, systems analysts find these methods are low cost, easy-to-implement alternatives. Included in

FIGURE 12.12 Selected forecasting techniques available to the systems analyst

this group are graphical extrapolation, moving averages, and analysis of time series data. Since these methods are simple, reliable, and cost effective, the remainder of the section focuses on them.

Analysis Of Time Series Data When quantitative historical data are present, the systems analyst can estimate future trends. The analyst should gather relevant data and graph it so as to determine if there are any patterns. There are three common patterns, as shown in

Figure 12.13: linear trend, seasonal, and cyclical. Fluctuating patterns are called seasonal trends if they repeat annually. If patterns repeat regularly but not annually, we call these patterns cyclical. The analyst needs to be aware of these patterns when estimating demand, workloads, and economic factors.

Estimation Of Trends Trends can be estimated in a number of different ways. The most widely used techniques are: (1) graphical judgment, (2) the method of least

FIGURE 12.13 Three factors that must be Identified for accurate forecasting

squares, and (3) the moving average method. A brief explanation of these techniques is in order. Graphical judgment. The simplest way to identify a trend and forecast future trends is by graphical judgment. This is accomplished by simply looking at the graph and estimating freehand an extension of a line or curve. An example of graphical judgment is illustrated in Figure 12.14. The disadvantages of this method are obvious from looking at the graphs in the figure. The extension of the line or curve may depend too

FIGURE 12.14 Different conclusions can easily be drawn from the same set of data.

much on individual judgment and may not represent the real situation. The graphical judgment method is useful however, because the ability to perform sensitivity analysis (what-if) has increased with the introduction of electronic spreadsheets The method of least squares. When a trend line is constructed, the actual data points will fall on either side of that line. The objective in estimating a trend using the least squares method is to find the “best fitting line” by minimizing the sum of the deviations from a line. Once the best fitting line is found, it can be graphed, and the line can be extended to forecast what will happen.

The best fitting line or least square line, is developed from the data points (X1, Y1), (X2, Y2),.. (Xx, Y), where the X coordinates signify the time periods and the Y coordinates represent the variable the systems analyst is trying to predict. The equation for the least square line is expressed in the form Y=m*X+b where the variable m represents the slope of the line and b represents the Y intercept, the point at which the line intersects the Y axis. An example of a least-square line is drawn in Figure 12.15. In this illustration, the least square line is drawn to estimate a trend in transact ions from January to October. The volume starts at 21,000 in January and rises an average of 2,000 units per month. We recommend a more computationally efficient method to find the least square equation by calculating the center of gravity of the data by taking x = X — X and v = Y — Y and then calculating the least square line as

 Σxy  y =  2 * x  Σx  finally substituting back the X — X for x and Y — Y for Y. Moving averages. The method of moving averages is useful because some seasonal, cyclical, or random patterns may be smoothed, leaving the trend pattern. The principle behind moving averages is to calculate the arithmetic mean or data from a group of periods

FIGURE 12.15 The objective of the least squares method Is to find the “best fittingline”

Y1 + Y2 + ... + YN N and then to calculate the next arithmetic mean by discarding the oldest period’s data and adding data from a new period

Y2 + Y1 + ... + YN −1 N and in this manner say the average is moving.

Figure 12.16 shows one type of moving average. Here five years data are averaged and the resulting figure is indicated. Notice that years 1976 through 1980 inclusive are averaged to represent 1978. Then the years 1977 through 1981 are averaged to get a representative figure for 1979, and so on. When this s graphed as it is in Figure 12.17, it is easily noticeable that the widely fluctuating data are smoothed. The moving average method is useful for its smoothing ability, but at the same time it has many disadvantages. First, the data at the very beginning and end is lost. In the above example, we lost data for 1976 and 1977 at the beginning and years 1985 and 1986 at the end. The trend line must be extended from 1984, rather than from 1986 as in least squares. Furthermore, moving averages are more strongly affected by extreme values than the methods of graphical judgment and least squares. There are many worthwhile forecasting packages available for microcomputers as well as for mainframes. The analyst should learn forecasting well, as it often provides information valuable in justifying the entire project.

FIGURE 12.16 Calculating a five-year moving average

Consulting Opportunity 12.2 The Birth of a System “Yup. What little there is all mine. I started in this business because could not keep my hands off of the stuff. I loved tinkering with our electronic equipment. Taking apart TVs and VCRs. Ask my wife. Lillian. Then I started helping friends with their projects, and they thought I was pretty good. When I inherited some money. I opened this little shop selling and repairing TVs and VCRs and renting videotapes” says D. W. Griffith, owner of a video rental and repair store. “Right now,” D. W. continues as he tows you around the small store, “we use a manual system for keeping track of rental videotapes. We make a 3 x 5 card for each title we own the name of the person renting the tape is recorded on this card.” “The cards were okay for a while, but now I need your help. I’m afraid that if things keep growing like they have. I’ll need a computer soon. But I didn’t get into this business overnight, and I want to be sure that the computer will pay for itself. I need proof. I’ve got to see It in black and white.” D. W. confides. ‘I was interested enough to keep a log of demand for rental tapes for the past 18 months.” D. W. continues. “Here ft Is.”

Hoping to catch up on some movies you’ve missed recently, you agree to do a small systems project for D. W. in return for a fee and some free videotaped films. Using the methods you have learned so far forecast the demand for tape rentals and the amount of new titles.

FIGURE 12.17 A five-year moving average smooth widely fluctuating data

Identifying Benefits And Costs Benefits and costs can be thought of as both tangible, and intangible. Both tangible and intangible benefits and costs must be taken into account when systems are considered.

Tangible Benefits Tangible benefits are advantages measurable in dollars that accrue to the organization through use of the information system. Examples of tangible benefits are an increase in speed of processing, getting otherwise inaccessible information, getting information on a more timely basis than possible before, taking advantage of the computer’s superior calculating power, and lowering the amount of employee time needed to complete specific tasks. And there are still others. Although measurement is not always easy, tangible benefits can actually be measured in terms of doll are saved, resources saved, or time saved.

Intangible Benefits Some benefits that accrue to the organization from use of the information system are difficult to measure but are important nonetheless. These are known as intangible benefits. intangible benefits include improving the decision-making process. enhancing accuracy. becoming more competitive in customer service, maintaining a good business Image, and increasing job satisfaction for employees by eliminating iedious tasks. As you can judge from the list given. intangible benefits u extremely important and can have far reaching Implications for the business in relating h people outside the organization as well as within it. While Intangible benefits of an information system are important factors In deciding whether to proceed with a system, a system built solely for Its Intangible benefits will not be successful, You must discuss both tangible and intangible benefits In

your proposal, since presenting both will allow decision makers In the business to make a well informed decision about the proposed system.

Consulting Opportunity The Birth of a System II: The Sequel “That’s a good question, I can think of a million problems we have here. but just off the top of my head. Here are the big ones” says D. W. Griffith during his second interview with you. its taking us lots of time to keep up records. It’s getting to be extremely difficult to sniff out tapes that aren’t returned. I like detective flicks, but I’m no Sherlock Holmes. I usually have to track down tapes myself. and that takes me away from my repair service. “What really gripes is about tapes that aren’t returned on time Is that they result In lost rentals, since most of my customers choose from the selections In stock that they can see on the shelves. And I don’t have any information about the categories of tapes my customers prefer. But part of it is my fault, I guess, since i haven’t bothered to set up a reservation system. I mean, there are probably more problems, but those are the ones I handle every day,” says U. W. You analyze the recording process and determine that employees can save two minutes for each tape rented and half a minute for each title in Inventory per month. Mr. Griffith himself gets involved with tracking down late tapes (about one for every 250 tapes rented) and can save half an hour per month (or each tape hi has to track down If he uses the computer Instead. Assume that Mr. Griffith can take out a small loan at 15 percent to purchase the computer and pay for development costs. The economy is seasonably good with an inflation rate of 5 percent Your time, as well as Mr. Griffith’s time, can be valued at £20 per hour. His employees time Is worth $5 per hour. You estimate that systems analysis will take thirty hours, system design forty hours, and development and Implementation thirty hours. Mr. Griffith will be involved In answering questions far about fifteen hours and in addition to the above you will have to individually train two people and Mr. Griffith for five hours each. The microcomputer, disk drives, printer, and software cost S4000, After the system Is up and running, you have advised Mr. Griffith that be will need to earmark $20 pm mouth for supplier and maintenance. After discussing your analysis with U. W. he says, “I don’t like giving you a lot of direction, but would you make a list for me? I’d like to screen all of this, so I can project what will happen. Give It your best shot. so I can take a dose-up look.” From the foregoing data, prepare a list of tangible and intangible costs and benefits for Mr. Griffith You will ha asked to analyze these costs In Consulting Opportunity 12,4.

Tangible Costs The concepts of tangible and intangible costs present a conceptual parallel to the tangible and intangible benefits discussed already. Tangible costs are (hose that can be accurately protected b the systems analyst and the business accounting personnel. Included in tangible costs are the cost of equipment such as computers and terminals, costs of resources, costs of systems analysts’ time, cost of programmers’ time, and other employees’ salaries. These Costs are typically well established or can be found out quite easily and are the costs that will require a cash outlay of the business.

Intangible Costs Intangible costs are difficult to estimate, and may not be known. They include the cost of losing a competitive edge; losing the reputation for being first with an innovation, or the leader in a field; declining company image due to increased customer dissatisfaction: and ineffective decision making due to untimely or inaccessible information. As you can imagine, it is next to impossible to project accurately a dollar amount for intangible costs. In order to aid decision makers who want to weigh the proposed system and all of its implications, you must include intangible costs, even though the’ are not quantifiable.

COMPARING COSTS AND BENEFITS There are many well-known techniques for comparing the costs and benefits of the proposed system. They include break-even analysis, pay-back, cash-flow analysis, and present value. All of these techniques provide straightforward ways of yielding information to decision makers about the worthiness of the proposed system.

Break-even Analysis By comparing costs alone, this kind of analysis allows the systems analyst to determine the break-even capacity of the proposed information system. The point at which total costs of the current system and proposed system intersect represents

the break-even point, where it becomes profitable for the business to get the new information system. Total costs include the costs that recur during operation of the system plus the developmental costs that occur only once (one-time costs of installing a new system)—.-that is, the tangible costs that were just discussed. Figure 12.18 is an example of break-even analysis on a small store that maintains inventory using a manual system. As volume rises, the costs of the manual system rise at an increasing rate. A new computer system would cost a substantial sum up front, but the incremental costs for higher volume would be rather small. The graph shows that the computer system would be cost-effective if the business sold about 600 units per week. Break-even analysis is useful when a business is growing and volume is a key variable in costs. One disadvantage of break-even analysis is that benefits are assumed to remain the same, regardless of which system

FIGURE 12.18 Break-even analysis for the proposed automated inventory system

is in place. From our study of tangible and intangible benefits, we know this is clearly not the case.

Payback Payback is a simple way to assess whether a business should invest in a proposed information system based on how long it will take for the benefits of the system to pay back the costs of developing it. Briefly, the payback method determines the number of years of operation that the information system needs to pay back the cost of investing in it. Figure 12.19 illustrates a system with a payback period of 3½ years. This is done through two ways—either by increasing revenues or increasing savings. A combination of the two methods can also be used.

FIGURE 12.19 Payback analysis showing a payback period of 3½ years

Since this is a popular way to assess alternative investments, businesses typically will have a set time period for payback assessments (three years, for example). This

is something you can find out from the accounting personnel with whom you are working on the systems project. If the proposed system has a projected payback of six years in a company that adheres to a three-year maximum payback on projects involving fast-changing technology, the system will be rejected. Payback that is made within the range used by the business, but is still longer than typical (i.e., four years instead of three), may not be rejected outright but may be subject to scrutiny through other methods. Although the payback method offers a well-known and simple way to assess the worthiness of the information system, it has three draw-backs that limit its usefulness. One drawback is that it is strictly a short-term approach to investment and replacement decisions; the second is that it does not consider the importance of how repayments are timed; and the third is that the payback method does not consider total returns from the proposed systems project that may go well beyond the payback year. Other forms of analysis should be used to augment the payback method and overcome some of these flaws.

Cash-flow Analysis Cash-flow analysis examines the direction, size, and pattern of cash flow that is associated with the proposed information system. If you are proposing the replacement of an old information system with a new one, and the new information system will not be generating any additional cash for the business, there are only cash outlays associated with the project. If this is the case, the new system cannot be justified on the basis of new revenues generated and must be examined closely for other tangible benefits if it is to be pursued further. Figure 12.20 shows a small company that is providing a mailing service to other small companies in the city. Revenue projections are that only $5,000 will be generated in the first quarter, but after the second quarter revenue will grow at a steady rate. Costs will be large in the first two quarters and then level off. Cash-flow analysis is used to determine when a company will begin to make a profit (in this case, it is the

FIGURE 12.20 Cash-flow analysis for the computerized mail addressing system

third quarter with a cash flow of $7,590) and when it will be “out of the red”—that is, when revenue has made up for the initial investment (in the first quarter of year 2 when accumulated cash-flow changes from a negative amount to a positive $10,720).

The proposed system should have increased revenues along with cash outlays. Then the size of cash flow must be analyzed along with the patterns of cash flow associated with the purc1ase of the new system. You must ask when cash outlays and revenues will be occurring, only for the initial purchase but over the life of the information system as well.

Present Value Present value analysis helps the systems analyst to present to business decision makers the time value of the investment in the information system as well as the funds flow (as discussed in the previous section). Present value is a way to assess all of the economic outlays and revenues of the information system over its economic life and to compare costs today with future costs and today’s benefits with future benefits. In Figure 12.21, system costs total $272,000 over six years and benefits total $280,700. Therefore, we might conclude that benefits outweigh the costs. However, benefits only started to surpass costs after the fourth rear, and dollars in the sixth ‘ear will not be equivalent to dollars in year 1. For instance, a dollar investment at 7 percent today will be worth S1.67 at the end of the year and will double in approximately ten years. The present value, therefore, is the cost or benefit measured in today’s dollars and.depends on the cost of money. The cost of money is the opportunity cost, or the rate that could be obtained if the money invested in the proposed system was invested in another (relatively safe) project. The present value of $1.00 at a discount rate of i is calculated by determining the factor

FIGURE 12.21 Without considert,ng present value, the benefits appear to outweigh the costs

FIGURE 12,22 Taking into account present value, the conclusion is that the costs are greater than the benefits. The discount rate, I, Is assumed to be .12 In calculating the multipliers in this table.

where n is the number of periods. Then the factor is multiplied by the dollar amount yielding present value, as shown in Figure 12.22. In this example, the cost of money

(i.e. discount rate) is assumed to be .12 (12 percent) for the entire planning horizon. Multipliers are calculated for each period: n = 1, n = 2, ... , a = 6. Present value of both costs and benefits are then calculated using these multipliers. When this is done, the total benefits (measured in today’s dollars) are $179,484—less than the costs (also measured in today’s dollars). The conclusion to be drawn is that the proposed system is not worthwhile if present value is considered. Although this example, which used present value factors, is useful in explaining the concept, all electronic spreadsheets have a built-in present value function. The analyst can directly compute present value in this manner.

Guidelines for Analysis The use of the above methods depends on the methods employed and accepted within the organization itself. However, for general guidelines it is safe to say: 1. Use break-even analysis if the project needs to be justified in terms of cost, not benefits: or if benefits do not substantially improve with the proposed system. 2. Use payback when the improved tangible benefits form a convincing argument for the proposed system. 3. Use cash-flow analysis when the project is expensive, relative to the size of the company; or when the business would be significantly affected by a large drain (even if temporary) on funds. 4. Use present value when the payback period is long, or when the cost of borrowing money is high. Whichever method is chosen, it is important that cost/benefit analysis be approached systematically, in a way that can be explained and justified to management, who will eventually decide whether to commit resources to the systems project. Next, we turn to the importance of comparing many systems alternatives.

Examining Alternative Systems Through the use of break-even, payback, cash-flow, and present value analyses, it is possible to compare alternatives for the information system. As was shown previously, it is important to use multiple analyses in order to cover adequately the shortcomings of each. Although you will consider several alternatives, the proposal itself will recommend only one. This means that you will have done comparative analyses about which system makes better economic sense before the proposal is

written. Those analyses can be included to provide support for the system you are recommending. Do not think that there is only one “correct’ system solution to help a business solve its problems and reach its goal. Different businesses call for different system attributes, and systems analysts themselves differ about the best way to handle various business problems. Based on your ascertainment of information requirements, the tangible and intangible costs and benefits of the system, and so on, compare Consulting Opportunity 12.4 The Birth of a System Ill: The Final Chapter “The list you made of tangible and intangible costs and benefits really helped.” raved D. W. “You had to do some tight editing to make anything out of that melodrama I spun for you the other day. Before we produce a system, however, I’d like you to analyze the tangible costs and benefits you wrote for me. Even films have critics.” D. W. reminds you, as he slips back out of view and heads back to his repair bench. Using one of the methods discussed in this chapter, perform the analysis D. W. called for, and make a convincing case for the systems solution you preview. Use graphs to get the proper angle on the project and put the debut of your analyses in its best light.

the alternatives with which you are working. This might call for the use of one of the multiple-criteria methods described in Chapter 11. Be open-minded as you compare alternatives so that the system you are recommending can be said to fit within the business as its members experience it. The key point is that you ‘want to compare and contrast options in as fair a manner as possible. so that a true choice is offered to organizational decision makers. The closer their initial identification with and acceptance of the proposed system, the greater the likelihood of continued use and acceptance once the system is in place. Continue including decision makers in the planning, even though you must in some ways expect to play the role of the systems expert now.

SUMMARY Evaluating hardware and software; identifying and forecasting costs and benefits; and performing cost/benefit analysis are all necessary activities the systems analyst

must accomplish in preparing material for the systems proposal. Information requirements help shape what software is purchased or written, as well as what hardware is necessary to perform required data transformation functions. Systems analysts must estimate workloads in order to adequately characterize current and projected workload capacity necessary for hardware. Sample workloads can then be run on hardware under consideration. Options for size and power of computer range from microcomputers to supercomputers. Systems require adequate data storage so that stored data are not endangered through periodic access to other data and so that users are not slowed by having to go through unstrored data during an inquiry. Many storage media are available: magnetic tapes are inexpensive and good for massive data storage; magnetic disks are good for on-line systems that require interactive capability; floppy diskettes are useful for smaller systems, but costly and time-consuming to back up; and hard disks provide large storage capacity for smaller computer systems. Although computer equipment changes rapidly, the process used in evaluating hardware need not. Through inventorying equipment already on hand and on order, systems analysts will be able to better determine if new, modified, or current computer hardware is to be recommended. Computer hardware can be acquired through purchase, lease, or rental. Vendors will supply support services such as preventive maintenance and user training that are typically negotiated separately. Packaged software must also be evaluated by the systems analyst and pertinent users. Much programming time can be saved if such a package is usable without extensive customizing. Software needs to be evaluated on how well it performs desired functions, its ease of use; adequacy of documentation: and other support services vendors may offer. Preparing a proposal means identifying all of the costs and benefits of a number of alternatives. The systems analyst has a number of methods available to forecast future costs, benefits, volumes of transactions, and economic variables that effect costs and benefits. Costs and benefits can be tangible (i.e., quantifiable) or intangible (i.e., nonquantifiable and resistant to direct comparison). A systems analyst has many methods for analyzing costs and benefits. Break-even analysis examines the cost of the existing system versus the cost of the proposed system. The payback method determines the length of time it will take before the new system is profitable. Cash-flow analysis is appropriate when it is critical to know the amount of cash outlays, while present value takes into consideration the cost of borrowing money. These tools help the analyst examine the alternatives at hand and make a well-researched recommendation in the systems proposal.