1 ON THE DISCOURSE OF MEASURING WORK FLOW EFFICIENCY IN CONSTRUCTION. A DETAILED WORK SAMPLING METHOD Bo Terje Kalsaas 1 ABSTRACT The purpose of this paper is to test out a detailed work sampling method in the operationalisation of workflow – as part of the discourse about workflow and waste. The paper is part of a wider, ongoing research project on flow in construction work, which has resulted in the identification of several alternative strategies for measuring workflow. The conceptualisation of workflow is based on the premise that it must integrate the two dimensions of uniformity (smoothness) and intensity (productivity). The method builds on an operationalisation of the factors involved in workflow which has its basis in Shingo’s well-know distinction between operation and process. Furthermore, the conceptualisation draws on the social methodology of critical realism in order to distinguish between different causes for good or bad flow. It also includes an understanding of waste as visible and observable, and as hidden in transformation work (direct work) and indirect work; and of what initiatives can be taken to reduce visible and hidden waste. An inductive approach is applied to operationalise operations in construction, and to the question of how this data based on work can be used to calculate workflow. The verification of the explored instrument or measurement model draws on data from different construction sites. Theoretically, the paper contributes to the Lean literature by conceptualising workflow and waste within a social scientific framework. Practically, it contributes by establishing detailed benchmark figures on the basis of different construction projects. KEY WORDS Workflow, waste, operationalisation, work sampling, measurement. INTRODUCTION This paper is part of an ongoing research project where the aim is to establish a method for measuring workflow in construction. In Kalsaas and Bølviken (2010) we sought to develop a conceptual description of flow that would enable its operationalisation – thus making is measurable – while retaining the concept’s intuitive qualities. The flow concept is understood in terms of two dimensions: throughput volume, and throughput uniformity. By uniformity is meant the degree of throughput volume stability per time unit. In short, the argument is that if we imagine a production process which delivers a completely stable, but minimal product volume per time unit, it makes little sense to describe this as a process that has a “good flow”. Conversely, it makes no more sense to describe a production process which delivers large volumes, but by fits and starts, as having a “good flow”. “Good flow”, then, is 1 Dr.Ing, Associate Professor, Faculty of economics and social sciences, Department of working life and innovation, University of Agder, 4846 Grimstad, Norway, Phone +47 37 25 30 00, e-mail: bo.t.kalsaas@uia.no