N. N. MITEV. The globalisation of transport? Computerised reservation systems at American Airlines and French Railways, In Rewiring Prometheus: History, Globalisation and Technology, edited by P. Lyth and H.
Trischler, introduction by Thomas Hughes, Aarhus University Press, Forthcoming, pp. 187-210, 2004.
The Globalisation of Transport?
Computerised Reservation Systems
at American Airlines and French Railways
Nathalie N. Mitev
Department of Information Systems
The London School of Economics, UK
Abstract
This case study examines the troubled introduction of a new computerised reservation system at French Railways. Socrate, based on the American Airlines Sabre system, had a disastrous beginning. It was badly received by the French public, led to strikes and government inquiries, and had to be modified substantially. The case study challenges beliefs and assumptions about the obvious success of technologies such as Sabre. The notion of 'symmetry' from the sociology of technology emphasises that failures express the same dynamics as successes, showing how technological choices are not obvious or unproblematic.
Differences between air and rail transport, between American and European transport deregulation and between the needs of national identity, regional development and public access to transport are all reflected in the question of yield management. Yield management is a crucial component of computerised reservation systems and was first adopted during the deregulation of the US air transport industry in the early 80s. It requires complex optimisation software designed to manage passenger revenues and control demand, by manipulating the availability of full and discounted fares according to monitored demand. Price differentiation is buried within the computer so that pricing, ticketing, and the choice of routes and trains are obscured to staff and customers.
The notion of 'translation' helps analyse how the Socrate project was undertaken and interpreted, as borrowing from airline pricing, aiming to gain competitive advantage, attempting to change passengers' buying and travelling behaviour, and helping identify profitable market segments. The Socrate case study exposes fundamental and controversial changes in transport. They are associated with the role of computer technology in deregulated and global markets, its effect on the concept of national identity and sovereignty in transport policy-making, and the relationship between global reservation travel systems and the future of European transport industries.
1. Introduction
1.1 Not just a failure
The interest of this case study is firstly as an implementation failure of a computer system in an organisation. Analysing the use of computer technology in organisations requires the appreciation of issues at micro-, organisational and macro-levels and of how they relate to each other. The stance taken here is that the usual notion of information technology failure is unhelpful. It belongs to a managerial and technicist discourse in which information technology (IT) is seen as unproblematic. Managers and computer practitioners tend to envisage IT as being neutral and evident, and these beliefs and assumptions of obviousness need to be questioned. Managerial and technicist discourses have a truncated understanding of organisations characterised by: a belief in 'rational' management; a denial of the continuing existence of power relations and conflict; a desire to eliminate organisational politics, for instance through the use of IT; a tendency to see organisations as individual closed entities, which have to survive in a hostile environment; and a limited focus on the business environment which neglects broader social, political, cultural, economic and historical perspectives. This narrow understanding of organisations colours the telling and explaining of IT failure and success stories. The explanations commonly found in the management literature are as simplistic for successes as they are for failures. In the case of failures they retrospectively try to find something/someone to blame, as if using information technology was a neutral, objective, rational exercise, which in the 'normal' course of event is successful and unproblematic. What is a more useful exercise is unearthing the accounts various groups make of these stories at different times and for what reasons; this should lead to a much richer, more complex, and less manichaean picture of the use of information systems in organisations. Our analysis must therefore move away from managerialist and technicist explanations, not limit itself to the organisation as a closed entity and include broader environmental perspectives.
1.2 The technical and the social
Common analyses of information systems failures [e.g. Flowers, 1996] reflect a convenient dichotomy between the technical and the social. Sociologists of science and technology have argued that the boundary between the social and the technical is a matter for social negotiation and represents no underlying distinction: "the fabric has no seams" [Bijker, Hughes and Pinch, 1989:11]. Technical and social choices are constantly negotiated and constructed, and their construction follows the same logic in successes and failures. Conversely, disentangling the interplay of actors and the construction of technical and social choices can be done by focusing on 'failure' studies as the latter can show that, at the beginning of and during a project, choices are not obvious or unproblematic, unlike what they appear to be in a successful project. Failure studies also allow for more elaborate explanations from actors. This leads to richer verbalisations of the complex links between technical choices and the social environment, and more readily so than in the case of 'successes' where choices tend to be seen as obvious, at least retrospectively.
"Like the sociology of science, the sociology of technology has chosen as its methodological principle to use the same explanatory resources when reporting on successful and unsuccessful innovations. However, to challenge the impression of obviousness which can be given by technical choices that lead to devices which 'perform well', there is no better strategy than concentrating on failure cases to show that it is impossible to distinguish between good and bad decisions. Moreover, in failures and controversial cases, actors facilitate the researcher's work since they express the more complex relationships between technical choice and social environment" [Akrich, 1993:36-37, my translation].
Historians and sociologists of technology [e.g. Bijker, Hughes and Pinch, 1989:14], borrowing from the sociology of scientific knowledge, argue that failures are of as much interest as the success stories, and that they should be explained 'symmetrically'. The sociology of scientific knowledge also recommends that scholars interested in the development of technology choose controversy as one important site for research. The controversy is about the truth or falsity of a belief, or about the success or failure of a technology, in solving social problems. Different groups will define not only the problem differently but also success or failure, and there is not just one possible way, or one best way, of designing an artefact [Bijker, Hughes and Pinch, 1989:14].
1.3 Sociology of technology
The history and sociology of technology also argue that the development of devices should be interpreted within an analysis of the struggles and growth of 'systems' or 'networks'. The constructivist approach to the study of technology moves away from making distinctions among technical, social, economic and political aspects of technological development and uses the 'seamless web', 'system' or 'actor-network' metaphors. There are various strands of social constructivism of technology, for instance, Callon who refuses to categorise the elements in a system or network "when these elements are permanently interacting, being associated, and being tested by the actors who innovate" [Callon, 1989:11]. Actor-network theory [Callon, 1986, 1991], a more extreme constructivist approach, uses a higher abstraction, 'actors', that subsumes science, technology, economics, politics. 'Human' and 'non-human' actors are the heterogeneous entities that constitute a network. By contrast, Bijker, Hughes and Pinch preserve the social environment. They argue that the social groups play a critical role in defining and solving the problems that arise during the development of an artefact. Problems are defined within the context of the meaning assigned by a social group or a combination of social groups. Because social groups define the problems of technological development, there is flexibility in the way things are designed, not one best way [Bijker, Hughes and Pinch, 1989:14].
1.4 Technology and globalisation
Setting aside failure issues and non-essentialist understandings of the role of technology, this case study is revealing of fundamental and controversial changes taking place in air and rail transport. The troubled implementation of an air computerised reservation system at French Railways is symptomatic of the difficult transformations caused by globalisation and the use of information and computer technologies, and relates to some of the themes covered in this book. French Railways adopted knowledge, information, technologies and management concepts developed by American Airlines in the 70s and 80s. Paradoxically, it was fascinated with air transport, which is related to the prestige of this technology in the 20th century, when compared to rail transport. Computerised reservation systems are information technologies with the potential to undermine the concept of national sovereignty. Global distribution systems (GDS) for travel are becoming increasingly transnational. Combined with transport deregulation, they are beginning to erode the role of the state and its responsibilities towards public transport and the transformation of economic progress into citizens’ welfare. There are tensions between transport systems perceived as national infrastructure technologies and how global computer technology is threatening national states and cultures.
To analyse the present case study, the notion of 'translation' will be used. According to Callon [1986, 1991] and Latour [1987], actors must have their attributes defined for them, or translated, so that they can play their assigned roles in the scenario conceived of by the socio-technical actor-network. In order for an actor to secure or win the support of others, it must make itself indispensable to them by translating their interests. The network becomes constructed according to the translations' own logic. After summarising the main events of the Socrate case study in Section 2, and outlining the surrounding economic and political context of European transport deregulation and the changes at SNCF in Section 3, Section 4 will attempt to apply the notion of translation to analyse the case study material. Finally in Section 5, the book themes will be discussed in relation to the case study.
2. The case study: 'modernising' French Railways
Fieldwork at SNCF (Société Nationale des Chemins de Fer Français) to investigate this case study began early 1994, six months after the problematic introduction of Socrate, and lasted eighteen months. Research access was relatively easy and the timing was judicious: access immediately after the events (summer 1993) would certainly have been refused as SNCF was greatly exposed to media scrutiny for several months and a climate of blame prevailed. I was also not seen as threatening, a French academic living in the UK, and most interviewees were very willing to talk at length to someone they perceived to be an outsider. In fact, they appeared to welcome an opportunity to re-examine events six months later, and many interviews lasted 3-4 hours. I was also given liberal access to files and documentation covering the initiation and development of the Socrate project, starting from 1989. Primary methods resulted in approximately 25 hours of audio-taped in depth individual and group interviews. Several of the original members of the Socrate executive team were interviewed, as well as SNCF senior managers, yield management experts, marketing, human relations and training managers, SNCF sales staff, railway union representatives, passengers’ associations, CRS experts and travel agents. Secondary sources included press reports, internal SNCF memos and documents, government and audit reports, technical documentation (pricing, ticketing, training manuals, software), unions’ and consultants’ reports. Material from interviews is used below and interviewees’ job titles during the Socrate project, and their organisations if other than SNCF, are indicated (in French).
2.1 The actors and the story
SNCF introduced Socrate (Système Offrant à la Clientèle des Réservations d'Affaires et de Tourisme en Europe), a computerised reservation system in April 1993. SNCF bought Sabre from American Airlines in 1989 in order to build Socrate. One of its aims was to transform its commercial activities through the instigation of a new philosophy of selling, based on a technological investment importing techniques used in the airline industry. One of the marketing slogans used by SNCF, "Avec la SNCF tout est possible"[1] [Naulleau, 1993] seems ironical since even the worst proved to be possible. Socrate is perhaps the first software system that provoked nation-wide strikes when it was introduced and which attracted massive negative media exposure. For such an ambitious project, and perhaps because of it, the number and type of problems encountered were rather spectacular: problems in its analysis and design, development and implementation, consultation, ergonomics, training, linked to a highly controversial commercial strategy and to communication blunders.
SNCF started the Socrate project in 1989. The aim was to reposition the enterprise in a new European competitive environment [Interview, Directeur Projet Socrate] that would involve a substantial traffic expansion. Socrate would offer a better quality of service, and support the diversification of services [Bentegeat, 1991]. The initiators of the project also explicitly emphasised the importance of an appropriate policy for maximising revenue, since SNCF had been an industrial and commercial establishment since 1982 (a semi-public as opposed to a strict public sector nationalised utility). One of their most important objectives was to instigate a new philosophy of selling based on yield management techniques [Interview, Responsable Informatique Socrate; see also Bromberger, 1993]. SNCF bought Sabre (initially SABER for Semi-Automatic Business Environment Research), the American Airlines computerised reservation system and classic example of an information system claimed to have provided competitive advantage to a major air company [Hopper, 1990]. Several years were spent adapting this software [Interview, Gestionnaire de Projet Socrate] developed by a private air company to the context of the rail industry and of a national semi-public sector institution.
When first implemented, however, SNCF staff and customers rejected this new technical tool and its underlying ticketing, pricing and selling policies. These implementation problems were widely reported and examined by SNCF itself [SNCF, 1993a; SNCF, 1993b], by French trade unions [CGT, 1994; FO, 1994], business consultants [APST, 1991; Causa Rerum, 1993], by passengers' associations [FNAUT, 1993], by the French government which commissioned a public inquiry into its implementation failure [Moissonnier, 1993; Cuq and Bussereau, 1994], and by the media [Faujas, 1993a; Pénicault and Riche, 1993; Henno, 1993]. Technical malfunctions, political pressure, poor management, unions and user resistance led to an inadequate and to some extent chaotic implementation [Eglizeau et al, 1996; Mitev, 1996]. The project management team gave a rather secondary importance to the databases and input sets [Interview, Gestionnaire de Projet Socrate]. Staff training was inadequate and did not prepare salespeople to face real-life problems such as tariff inconsistencies and printing problems [Interview, Manager de la Formation Socrate]. The user interface was designed using the airlines logic and therefore was not at all user-friendly. The new ticket proved unacceptable to customers. Public relations failed to prepare the public to such a dramatic change [Interview, Délégués aux Missions Extérieures]. The inadequate database information on timetable and routes of trains, inaccurate tariff information, and unavailability of ticket exchange capabilities caused major problems for the SNCF sales force and customers alike. Impossible reservations on some trains, inappropriate tariff and wrong train connections led to large queues of irate customers in all major stations and to a major public outcry in France. Online reservations available through the Minitel public network failed, booked tickets were for non-existent trains whilst other trains ran empty, railway unions went on strike [Devillechabrolle, 1993], and passengers' associations sued SNCF [D'Aufresnes, 1993; Faujas, 1993b].