Framework for Ensuring Risk Assessment Completeness In Construction Industry Abel Pinto 1 , Isabel L. Nunes 1 , Rita A. Ribeiro 2 1 Universidade Nova Lisboa/FCT, Caparica, 2829-516 Portugal 2 Uninova, Campus UNL/FCT, Caparica, 2829-516 Portugal Email (contact): abel.fnpinto@gmail.com In work environments, risk assessment is essential to prevent injuries and health problems. This created the necessity to identify criteria for evaluating the results of such analysis, in terms of completeness and depth of assessment. In the construction industry there are, perhaps more than in others, important issues to take into account to conduct a proper RA. The aim of this paper is to present a checklist framework to evaluate the RA process completeness and adequacy of results, in terms of quality (fitness for use) in the construction industry. Specifically, we will address questions related with hazards and respective contributing factors; accuracy of risk assessment; preventive or protective measures needed; available data and information; safety barriers against damages and work injuries. INTRODUCTION In the construction industry, risk assessment (RA) is compulsory (legal obligation) in all developed countries, however, accidents still happen every day. RA is performed to document that the risks associated with a given activity are (or not) acceptable. To the best of our knowledge, there aren‟t any International Standards for RA process in the occupational safety and health context; except RA for technological systems (Rouhiainen, 2002). Due to its importance for support safety programs, it is essential to have methods and criteria to evaluate the RA process to detect its deficiencies. According NP EN ISO 9000:2005, quality means degree of satisfaction of requirements (needs or expectations). The aim of RA is to identify essential factors affecting the safety (and workers health) of an activity, equipment or site and to decide which safety measures should be taken (Ringdahl, 2001). So, in this context, RA quality means the degree to which RA is suitable for establishing, correctly, a safety program on site. The most important aspects for low quality in RA results are: limitations of hazard identification methodologies; inaccuracies in available data; lack of identification of consequences; lack of organizational and human data. Some authors (Suokas, 1985; Taylor, 1982, cited in Rouhiainen, 1992) pinpoint that the process of RA depends, greatly, on the analyst or the analysis team. Differences are mainly caused by:  The initial assumptions;  Restrictions on the object or system [the exposed humans, environments and/or physical objects (Christensen, 2003)]);  Lack of information;  Failures and human errors referring to the data employed, and;  Problems in the estimation of consequences. Henceforth, we use system to describe the exposed humans, environments and/or physical objects under analysis. Construction projects are becoming increasingly complex and dynamic in their nature; hence to perform risk assessment is a complex task and subject to high imprecision and uncertainty (Carr &Tah, 2001). An appropriate RA is the basis for a successful risk assessment; if the analysis does not have quality, the risk assessment is obviously inadequate. Some authors (Rouhiainen, 1992, Ringdahl, 2001) proposed some recommendations to assess the RA process quality. Measurements of RA performance (in terms of completeness and depth of assessment), requires quality criteria and performance metrics to determine its adequacy of completion (Palmer, 2004). Since there are not many contributions for identification of quality criteria and suitable measurements of RA results, it is important to create a RA quality framework.