Software failures have a habit of making the news: the more spectacular the failure, the larger the headline. While the front pages of the professional computing journals are normally reserved for such stories, failure stories also have a tendency of making national and even international headlines. The statistics on information systems/information technology (IS/IT) failure are quite telling. Current estimates suggest that the annual cost of IS failures in Western Europe adds up to US$140 billion, while the USA is reckoned to lose US$150 billion in any given year. With such alarming figures, individual failure stories are often compared to tips of icebergs, thus representing but a mere fraction of the true scale of the problem. One recent example of a high profile implementation failure is related to the troubled Libra project in the UK. 1 Libra was designed to provide a standard IT system for magistrates’ courts including upgraded infrastructure, office automation facilities, a national casework application and electronic links with other criminal justice agencies. The original contract for £184 million was awarded in 1998 but following implementation problems the deal collapsed in July 2002. Following re-negotiation later in 2002, the infrastructure portion alone cost £232 million. The total system would now take 8.5 years to develop and cost over £318 million. The chairman of the House of Commons’ Public Accounts Committee, Edward Leigh, described the system as one of the worst IT projects he had come across, especially as it was likely to cost more than twice the original estimate and would now be available (to deliver benefit) for two years less than anticipated. From a purely economic perspective, a product (or an IT system) will be deemed successful if the returns on any investment needed exceed the costs of development as well as any other limitations that may ensue during or following the development effort. The cost is primarily determined by the finance expended on development. However, delay to the utilization of some of the benefits can also count as a cost. The utility derived from a system encapsulates the perceived returns generated by the system and its usage. Failures can be quantified in different ways, however, so that money wasted on a project is not the only criterion that defines ‘cost’. Certain projects are likely to have an impact beyond the initially predicted boundaries, and may even lead to losses elsewhere within the system. Another recent disaster in the UK, related to an earlier implementation failure concerns the delay in introducing the Nirs2 system into the Inland Revenue beginning in 1995. This resulted in backlogs that caused the Inland Revenue to stop sending reminders to up to a third of the UK working force warning them that they
Search from 250 M+ research papersSearch from 250 M+ research papers
Open Access
Dec 31, 2012
You-Jon Won + 2
