Home Random Page


CATEGORIES:

BiologyChemistryConstructionCultureEcologyEconomyElectronicsFinanceGeographyHistoryInformaticsLawMathematicsMechanicsMedicineOtherPedagogyPhilosophyPhysicsPolicyPsychologySociologySportTourism






Information Systems Failures.

So far we have introduced you to many success stories. You may wonder, though, is IT always successful? The answer is, “Absolutely not.” There are many failures. We will show you some of these (marked with a “lessons from failures” icon) in this book, and in some cases we present them on our Web site. (See, for example, the 2000 U.S. Presidential Election Case in Online File W!.8 at the Web site.)We can learn from failures as much as we can learn from successes, as illustrated in IT at Work 1.2 (page 26)

One area of IT failure is that of the dot-coms. As will be seen in Chapter 4, hundreds of dot-coms folded in 2000 and 2001. It was a shakeout that resulted from a rush to capitalize on e-commerce (see Kaplan, 2002). In addition there were many failures of Internet projects in established companies. (For example, the Go.com project of Walt Disney Company was supposed to manage all the Web sites owned by Disney Company was supposed to manage all the Web sites owned by Disney and generate money from advertisers at the sites. Unfortunately, the income generated from advertising was not sufficient to keep the site going.)Like the gold rush and the rush to create companies when the automobile was invented, only a relatively few made it. The rest failed. According to Barva et al. (2001), the reason for EC failures is that many of the models used were too narrow. In place of these models, they offer an e-business value model, which we describe in Chapter 4.

Another reason for failure is that it is hard to predict the future. It is especially hard to predict the future in the field of information technology, which is evolving and continuously changing, as shown in Section 1.4.

IT at Work

How Nike’s $400 million supply chain

Management’s software system failed

 

In certain retail stores, fans of Nike’s Air Terra Humara 2 running shoe hit the jackpot. Once selling for over $100 US, they were selling for less than $50 in fall 2001. The cheaper shoes were the aftermath of the breakdown in Nike’s supply chain, a breakdown attributed to a software problem.

Nike had installed a $400 million supply chain system in early 2001. The system was supposed to forecast sales demand and plan supplies of raw materials and finished products accordingly. However, the newly deployed demand and supply planning application apparently overestimated the demand for certain shoes in some locations and underestimated demand in others. As a result, some raw materials were overpurchased, while inventory levels of other materials were insufficient. Some shoes were overmanufactured, while the most –demanded ones were undermanufactured. To speed the right shoes to market, Nike had to spend around $5 a pair in air freight cost, compared to the usual cost of 75 cents by ocean shipping. In all, Nike attributed some $100 million in lost sales in the third quarter of 2001 alone to this problem.

What went wrong? The system was developed with software from i2, a major software producer. However, Nike insisted on modifying the i2 standart software, customizing it to its needs. Specifically, Nike wanted a forecast by style, by color, and by size (several thousand combinations). This resulted in a need to make thousands of forecasts, very rapidly, to quickly respond to changing market conditions and consumer preferences. To meet Nike’s need it was necessary to customize the standard software, and to do so quickly because Nike wanted the system fast. The software had bugs in it when it was deployed. Almost any new software contains bugs that need to be fixed; appropriate testing is critical, and it is a time-consuming task (see Murphy, 2003). Nike and i2 failed to recognize what was achievable.



Customizing standard software requires a step-by-step systematic process (see Technology Guide 6). It should be done only when it is absolutely necessary, and it must be planned for properly. Furthermore, Nike could have discovered the problem early enough if they had used appropriate deployment procedures (see chapter 14).

To avoid disasters such as the one Nike experienced, companies must fully understand what they are trying to achieve and why. They must use performance-level indicators to properly measure the system during testing. Incidentally, Nike fixed the problem after spending an undisclosed amount of time and money in 2002.

Sources: Compiles from Sterlicchi and Wales (2001) and from nike.com press releases, 2002, 2003.

For Further Exploration: Why did Nike need the detailed forecasting? How can a company determine if it really needs to customize software? Whose responsibility is it to test and deploy the software: the software vendor’s or the users?

 

Information Technology Developments and Trends

In the previous sections, we described the role of IT in supporting business activities . We also pointed out (in Table 1.1, page 5) some of the capabilities that enable IT to play a support role. Next we will describe some of its developments and trends, and especially the move toward Web-based computing, wireless applications, and intelligent systems.

First imagine this scenario: It’s a Monday morning in the year 2008.Executive Joanne Smith gets into her car, and her voice activates a wireless telecommunications-access workstation. She requests that all open and pending voice and mail messages, as well as her schedule for the day, be transmitted to her car. The office workstation consolidates these items from home and office databases. The message-ordering “knowbot” (knowledge robot), which is an enhanced e-mail messaging system, delivers the accumulated messages (in the order she prefers) to the voice and data

T at Work

How Nike’s $400 million supply chain

Management’s software system failed


In certain retail stores, fans of Nike’s Air Terra Humara 2 running shoe hit the jackpot. Once selling for over $100 US, they were selling for less than $50 in fall 2001. The cheaper shoes were the aftermath of the breakdown in Nike’s supply chain, a breakdown attributed to a software problem.

Nike had installed a $400 million supply chain system in early 2001. The system was supposed to forecast sales demand and plan supplies of raw materials and finished products accordingly. However, the newly deployed demand and supply planning application apparently overestimated the demand for certain shoes in some locations and underestimated demand in others. As a result, some raw materials were over purchased, while inventory levels of other materials were insufficient. Some shoes were over manufactured, while the most –demanded ones were under manufactured. To speed the right shoes to market, Nike had to spend around $5 a pair in air freight cost, compared to the usual cost of 75 cents by ocean shipping. In all, Nike attributed some $100 million in lost sales in the third quarter of 2001 alone to this problem.

What went wrong? The system was developed with software from i2, a major software producer. However, Nike insisted on modifying the i2 standard software, customizing it to its needs. Specifically, Nike wanted a forecast by style, by color, and by size (several thousand combinations). This resulted in a need to make thousands of forecasts, very rapidly, to quickly respond to changing market conditions and consumer preferences. To meet Nike’s need it was necessary to customize the standard software, and to do so quickly because Nike wanted the system fast. The software had bugs in it when it was deployed. Almost any new software contains bugs that need to be fixed; appropriate testing is critical, and it is a time-consuming task (see Murphy, 2003). Nike and i2 failed to recognize what was achievable.



Customizing standard software requires a step-by-step systematic process (see Technology Guide 6). It should be done only when it is absolutely necessary, and it must be planned for properly. Furthermore, Nike could have discovered the problem early enough if they had used appropriate deployment procedures (see chapter 14).

To avoid disasters such as the one Nike experienced, companies must fully understand what they are trying to achieve and why. They must use performance-level indicators to properly measure the system during testing. Incidentally, Nike fixed the problem after spending an undisclosed amount of time and money in 2002.

Sources: Compiles from Sterlicchi and Wales (2001) and from nike.com press releases, 2002, 2003.

For Further Exploration: Why did Nike need the detailed forecasting? How can a company determine if it really needs to customize software? Whose responsibility is it to test and deploy the software: the software vendor’s or the users?



Date: 2015-12-18; view: 608


<== previous page | next page ==>
WHAT IS A COMPUTER-BASED INFORMATION SYSTEM? | List of participating bowlers
doclecture.net - lectures - 2014-2021 year. Copyright infringement or personal data (0.002 sec.)