Business Research Environment of Confirm Computerized Reservation System - Case Study Example

Report on Confirm – Computerized Reservation System Introduction Confirm Computerized Reservation System derives its roots of origin from the legendary American Airlines Computer Reservation System “SABRE” (Semi Automated Business Research Environment). The SABRE System was an initiative to achieve a centralized system to offer online reservations/bookings at competitive rates by examining the fares offered by competitors and setting competitive prices. In this way, SABRE also provided maximum profitability to American Airlines and help to reach the top notch position as the largest share holder in the US Airlines market by beating its competitor, United Airlines. In 1988, SABRE Airline Solutions launched Software, Consultancy and Systems Management Services to other airlines in areas such as Revenue accounting, Yield management and Crew scheduling. In 1988, The SABRE System expanded to store 36 million fares which could be combined to create more than one billion fare options. Case Study - Confirm Computerized Reservation System In March 1988, there arose a dire need of an online reservation system to cater to the bookings of hotel rooms and rental cars of some of the largest hotel and travel companies in the US. Hilton Hotels Corp., Marriott Corp., Budget Rent a Car and AMR Information Services had partnered to develop a System like SABRE that was called “CONFIRM” to manage the online booking of hotel rooms and rental cars. The proposed system was designed keeping in mind the voluminous data regarding the various hotels and rental cars information and details of all services provided to the users. The System will deal with providing details of availability of Room and Car for booking with information of any special offers/discounts available, along with an efficient method of making the bookings. For the hotel or car rental company, Confirm will provide information as to how to defend its market share, build detailed profiles of customer preferences and also provide the opportunity of achieving gains in efficiency by using Confirm’s yield management system. Confirm was designed like SABRE with a centralized IBM mainframe system which could store the massive amount of data hosted by hotels and car rental companies. Confirm was designed for multiple users i.e. travel agents, corporate travel departments and a large volume of individual users via workstations that are linked to existing reservation systems. The Confirm Project was a joint development venture by AMRIS (AMR Information System), Hilton, Marriot and Budget, with the work allotted to a unique entity “ABHM Partnership”, that was called “International Reservations and Information Consortium (Intrico) partnership” in later days. AMRIS played a twin role acting as partner and primary developer on this joint venture. The Intrico board was to manage the development process. In 1988, AMRIS, Hilton, Marriot and Budget combined together to integrate Airlines, hotel and car rental services. Once the project was ready, the goal was to market the project to the world. Confirm was designed by merging two different processes running on IBM linked 3090 mainframes. One of these computers, running IBM’s Transaction Processing Facility operating system was designed to manage the central reservation system, to be coded in C language and at the same time maintain records of Hotel room, Car status and availability. The second mainframe, running under the MVS operating system, was designed to handle the DB2 relational databases which will store information on pricing, customer histories and marketing analysis. The key component of the system, the TMF, a suite of applications developed using IEF case tool was designed to be the main link between the two systems and the users. The TMF program was designed to handle the interface with five major airline reservation systems, provide links to a numerous hotel management systems and all single users, and also coordinate network traffic. The TMF was the key component of the Confirm System. The budget for the Confirm development was £83million. Failure reasons/risk factors identified AMRIS played the key role in development and put forward the base design of the System but the other members could not make out the specifications of this system. The reasons of dispute between the different groups of the Joint venture according to the various news paper reports are: 1. Definition of System functionality 2. Missed deadlines 3. Request to Modify Specification 4. Voluminous number of change requests in the system 5. Reporting procedures for technical and scheduling problems 6. Project management skills not put to proper use 7. The developers were not able to create a fault-tolerant database, which is a critical requirement of the system 8. Choice and usage of development tools. The Case tool was unable to integrate the two major components within Confirm. 9. Lack of proper communication among the different partners 10. Suitability of managers from all participants involved (knowledge of existing system and development) The project review done by IBM in 1989 underlined some key issues in project management areas such as 1. Intrico has to design a common agreeable goal to all partners, 2. Intrico has to adhere strictly to time management and alert all the members before hand when deadlines are missed, 3. Intrico has to define effective reporting structures to show the key details of a project’s progress, 4. Intrico has to perform regular audits to assess system competence as well as users, 5. Intrico has to ensure proper communication between all its members, 6. Intrico has to allocate technically knowledgeable project developers for the project. In 1992, the Confirm system was ready to be put for -testing for Hilton Hotels Corp. The technical experts from SABRE team audited the Confirm system and identified the key issues as 1. The two subsystems on different mainframes relied upon Transaction management function (TMF) to act as a bridge between them but this link was interactive with very slow response time and much of the information was not accessible through this link. 2. Another key issue was under-estimation of the complexity of providing application to application link between mainframe processors for 60 applications. 3. The decision of Intrico to use a CASE tool to automate generation of code while writing the TMF subsystem was faulty. The TMF did not work as per the expectation. 4. At the advent of system crash, the Confirm DB2 database lacked the ability to restart and continue operations. The reason behind this was because IEF case tool was allowed to generate its own database structure. The automatic code generation is usually long winding and requires lot of modifications. Once the key areas of issues were identified, Staff review was conducted the result of it being twenty Information Systems Managers were reassigned and further eight managers resigned. It was heard that employees were gripped with fear and could not discuss their issues with the Management. Due to these major issues, the deadline was to be extended by eighteen months further and implementation to be delayed by one year. In the middle of 1992, all the partners Hilton, Marriot and Budget dropped out of the Project. AMRIS continued for a few weeks more before it also, dropped out of this project. The legal disputes went on between them and finally they had settled the dispute among themselves, out of court and hence no information is available to the public as to how it got resolved. What could have been done to prevent failure? Confirm CRS could not take off as expected due to the major issues as mentioned. Now, the major issues that were to be addressed was 1. Before designing complicated application software, the key users have to interact together and write the specifications of the proposed system so that there are no further modifications and all key issues being addressed properly leaving little room for revision or change. 2. The missing of deadlines issue is to be addressed by concentrating on the developers’ team and solving their issues than by increasing the manpower as done in this case. The increase in manpower creates additional issues for the Management in terms of additional recruitment, training and extension of time until the new team members understand the core process. 3. Project Management skills were not put to use effectively by the Management in the sense, common addressable goal not System was not subject to technical audits, reporting procedures to be created to monitor the progress of the project, Communication between the partners has to be duly maintained and the major issue is to provide a technically strong development team for the project. 4. This project was scheduled to be completed in 1992, with the estimated cost $55.7 million and anticipated processing cost less than $1.05 per transaction. 5. According to Dennis Hart, “Functional and technical specifications were declared to be complete in early 1989 and design by late 1989. Price rose to $72.6 million in late 1989. The lapse of deadline created Management and Personnel issues. The price increased to $92 million in 1991 and led to resignation of Staff. The actual project status issues were not let out by the Management. In 1992, after expenditure of $125 million, the project was dropped. There were legal disputes between the concerned parties finally led to out of court settlement among them in 1994.” Lessons learnt The lessons learnt from the failure of Confirm were 1. It is easier to design a strategic information system but to take it to successful implementation levels is difficult. 2. Specifications are to be done prior to development process unlike this case where innumerable requests were pooled to alter the specifications now and then. 3. Code generation allowing automation of database structure was a lapse in the project. Database structure design has to be in control of the development team. Automation results in lengthy generations and needs modification but future changes become very complicated. 4. Project execution did not happen as per fixed deadlines and so the exorbitant increase of costs with time led to the failure, rather collapse of the entire system. Adherence to time lines is an important criterion for the success of a major project like Confirm. 5. Employee motivation was lacking in this Project. Many of the employees left the organization in the middle of the project development and so implementation was further delayed. Management has to address the issues of all employees in an efficient manner which is useful in holding the employees in the organization and successfully deliver projects. 6. The Project management skills include allocating of technically well equipped personnel for the Coding process. In this project, technically sound professionals could have built a fault tolerant database and develop proper interface between the database and user and enable multi-processing, transaction management and develop bug-free coding. Conclusion The proposed system Confirm CRS was an ideal multifaceted system designed to cater to three different industries, viz. Airline/Hotel/Car. The major milestones of the Project could have been met, if criteria specified in the report are adhered. This system study is useful for Information System students/Software novices, helpful to avoid the pitfalls in complicated System design/development/implementation. List of references 1. http://www.scit.wlv.ac.uk/~cm1984/qmp/failures/FOUR.HTM case study: CONFIRM 2. http://www.academia-research.com/files/instr/300106_IMG00031-20090513-2309.jpg 3. Dennis, Hart. www.teaching.fec.anu.edu.au 2006 16 May 2009 . Read More