From the perspective of service economics, two series of problems appear when evaluating the performance of services:
When applied to the case of freight transport, the conceptual framework of service economics allows us to specify the service relation between shippers and transport firms by describing the variety of freight transport demand, and to specify the “product” of transport in terms of short term output and long term ‘outcome’.
Transport services defined in the framework of specific service relations
The definition and delimitation of activities creates a first series of analytical problems. As a matter of fact, freight transport can be internalised in an industrial activity, or an externalised service activity. If we define freight transport as a service activity, we can turn to the service literature which has attempted to characterize the specificities of services in general.
According to the standard technical definition, which can be traced back to the classical view of “unproductive services” (Smith, Marx), a service is considered as immaterial, co-produced between user and producer, non-storable, and non-transportable.
Hill  criticizes the distinction between material goods and “intangible” services. Defining services by their intangibility introduces confusion, and the category of intangibles should be recognized as a type of good. Hill thus suggests making a distinction between three categories: tangible goods, intangible goods, and services. The two essential characteristics of services are that they “…cannot be produced without the agreement, co-operation and possibly active participation of the consuming unit” and that “the outputs produced are not separate entities that exist independently of the producers or consumers” (, p. 128).
Following Hill , the definition of services was further developed in a more socio-technical approach, focusing on the concept of “service relation”. A service relation is considered as a particular social relation between a producer and a user, leading to a change in the status of a reality owned by the user. Service activities can thus be defined as operations aimed at the transformation of the state of a reality C, performed by a service provider A for a user or client B, the result of which is not an independent product which can circulate separately from the reality C .
The concept of ‘service relation’ thus makes it difficult to identify a ‘standard service’ in freight transport. The physical transfer of goods in space (measured in terms of ton-km) is only one aspect of a service relation which is specific to a shipper and a carrier.
We can easily observe that freight transport operations have become more and more complex and differentiated over the past 30 years. Beyond shipping and handling goods, they more and more often include operations such as the treatment of information flows, the differentiation of goods for the final customer etc. The conceptual representation of freight transport exclusively in terms of flows of goods thus becomes less and less relevant to the realities of the freight transport sector and its performance issues. A more relevant conceptual representation of the ‘product’ of freight transport must include these different operations [30, 31].
Nevertheless, even in the service economics literature, freight transport is generally considered as a standard, simple and material service, that is to say having a material object. Following Gadrey’s  definition of services, different objects of services can be distinguished: material objects, people, or information. According to this typology of services, freight transport clearly belongs to the first category, moving material objects in space.
Similarly, in Du Tertre’s [15, 16] typology of production systems (“configurations productives”), freight transport would belong to logistic services, the productivity of which is mainly determined by direct intensity of labor, scale economies and material integration.
Djellal  discusses the distinction between services dealing with information and “non-informational” services and demonstrates, in particular for road haulage services, the co-existence of material configurations with a growing integration of information, methodological and relational aspects. The introduction of information and communication technologies (ICT) has transformed the nature of these services [1, 13, 21, 23, 28].
In a previous paper , we have discussed the co-existence of different types of operations in freight transport services. Following Gadrey  and Gallouj , we distinguish four types of operations in freight transport: material operations (the basic object of transport), treatment of (codified) information, relational or contact operations, and methodological operations (Table 1).
Material operations (M) concern mainly the traditional purpose of freight transport: shipping objects from one point in space to another, as well as handling of goods, loading and unloading. More recently, however, transport service providers sometimes provide other material services such as warehousing, packaging and labeling.
The diversity of potential combinations illustrates the fact that transport services, as all types of services, are heterogeneous products. Moreover, the array of combinations changes over time, which makes it impossible to analyse the dynamics of the freight transport sector through the measurement of a standard service.
Even the most orthodox analyses in transport economics admit that ton-km are an obsolete measurement of the output of freight transport. Quite obviously, transport activities do not produce weights nor distances. Ton-km only measure the physical labour of transferring goods in space, which was probably the major preoccupation of carriers at the beginning of the 20th century, when 90 % of goods were transported on rail and waterways. According to Fritsch and Prud’Homme (, p. 7), «the ancient and wide-spread habit of using ton-km continues in spite of the diversification of transport modes and transport services».
As a matter of fact, the ton-km remains, at the same time, the most widely criticized and most wide-spread indicator of output of the freight transport sector, which leads to an erroneous representation of freight transport performance. A correct and robust representation of ‘output’ of freight transport should take into account all the relevant service operations performed on the goods forwarded.
Transport is a process : the importance of the temporal dimension
If services produce a ‘change of state’, it becomes difficult to distinguish between product and process. Moreover, the temporal dimension of the change of state produced by a service is crucial. The more recent service economics literature distinguishes between ‘output’ and ‘outcome’ or ‘immediate’ outputs and ‘mediate’ outputs . In the health sector, for example, indicators of output could be the number of medical acts performed, the number of days of hospitalization etc. whereas the outcome could be measured by a change in life expectancy; similarly, educational services can be evaluated by their output (number of students, number of teachers of teaching hours), or by their outcome, for instance the level of education in a given population.
This distinction between output and outcome can be applied to freight transport services. The output of freight transport simply concerns the goods transported from one point in space to another. The outcome of freight transport services, however, goes far beyond this aspect of moving goods in space and concerns its contribution to an efficient supply chain on the whole, in terms of constant adaptation to the production system, just-in-time delivery, flexibility, reliability and input to the information and management system of the whole supply chain.
The usual productivity indicators based on ton-km thus only evaluate the immediate output of freight transport services. On the other hand, the outcome of these services, which is related to the efficiency of the articulation between the freight transport system and the production system for the goods to be forwarded, is not measured in the traditional indicators. In order to evaluate long-term performance, it is necessary to go beyond the single quantitative indicator and to develop a multicriteria framework taking into account the qualitative aspects of performance.
Fritsch and Prudhomme (, p. 17) recognize the difficulty to take into account the “quality of service” of freight transport. We think, on the contrary, that the dimensions of quality of service can be defined and are even quite well known—just-in time delivery, reliability, safety, sufficient production capacity (cf. Fig. 1 above)—but that the problem of evaluation is related to the restriction of the evaluation framework to a single quantified indicator (such as the ton-km).
The analysis of freight transport performance should be related to the industrial production system in which the transport service takes place. In the management literature, logistic performance is more and more often measured in terms of customer satisfaction [32, 34]—not simply from the point of view of the shipper, but with regard to the global supply chain. Freight transport performance indicators should encompass the global efficiency of flows of goods and information inside a production system. In such a framework, the main levers to improve performance are the qualitative and quantitative adaptation to production systems instead of mere cost reduction and investments in vehicle technology or infrastructures.
The main source of performance of freight transport services lies in their articulation with the production system
This analytical framework also sheds a different light on the recent evolution of freight transport and logistics in France. The rapid development of road haulage compared to the “heavy” modes of freight transport cannot only be explained by its productivity, nor by regulatory and technical constraints, as it is often debated in France. The type of transport services that this mode delivers appears to be particularly well fitted to the organizational and market structure changes that began in the 1970s. In an environment of rising differentiation of consumption goods, shortened life cycles of products and just-in-time production, logistic management systems become a strategic function in production systems. In this context, road haulage becomes the most efficient mode of transport for frequent, rapid and flexible deliveries, in comparison to railroads and waterways, far more competitive for large-scale transport flows of heavy or bulk goods. Trucks became, in a way, the symbol of the post-fordist accumulation regime.
In the 1990s, the structural changes in production and distribution methods lead to a reorganization of logistic and transport systems, in particular through the development of global supply chain management, involving producers, large-scale retailers, and logistic integrators in collaborative logistic processes. Freight transport services become part of more and more complex logistic processes, and transport firms cover a large variety of situations, from third party logistic providers who also provide transport services, to highly specialized companies, or very small scale subcontracting firms.
The adoption and development of information and communication technologies in freight transport, such as EDI (electronic data interchange) and web EDI, GPS, RFID (radio-frequency identification) etc., and the development of new methods of organizing logistics, information and transport flows (tracking and tracing methods, cross docking hubs, etc.) transform logistic processes in general and freight transport services in particular.
As a result, freight transport services, and in particular road haulage services, are nowadays far more complex than transporting a good from point A to point B. We will illustrate this point in the second part of the paper.
To sum up, the traditional productivity indicators used in transport analyses, mostly based on the ton-km as a measure of output, only evaluate a direct, short term output. Firstly, these indicators only include a specific type of service—transfer of goods from one point to another in space—and exclude all other service operations that have become more and more important in the freight transport industry in the recent past. Secondly, freight transport produces other results on the long run, and in particular important impacts on the efficiency and long term evolution of industrial production systems which are simply not accounted for in quantitative productivity indicators, although they become more and more critical in terms of sustainability of the systems of production and circulation of goods.
In the second part of the paper, we will illustrate and apply a more comprehensive and qualitative framework of evaluation of service performance.