ScienceDirect IFAC-PapersOnLine 48-3 (2015) 106–111 Available online at www.sciencedirect.com 2405-8963 © 2015, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved. Peer review under responsibility of International Federation of Automatic Control. 10.1016/j.ifacol.2015.06.066 © 2015, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved. 1. INTRODUCTION Over the past 25 years, enduring empirical evidence demonstrates the significant impact of Business Process Reengineering (BPR) on organizational performance (Altinkemer, Ozcelik, & Ozdemir, 2011). Well beyond firm- level performance, research has also shown how Supply Chain Management (SCM) practices require BPR within an operations strategy and inter-organizational benchmarking perspective (Herzog, Tonchia, & Polajnar, 2009). Typical BPR case studies address highly complex projects where readiness assessments clearly show prior challenges, such as organizational resistance or lack of project alignment with strategy (Abdolvand, Albadvi, & Ferdowsi, 2008). Yet even in the most favourable conditions, supply-chain reengineering may face the possibility that project plans are entirely flawed, where projects transition from optimistic inception and elaboration phases, to failed redesign and testing phases (Guimaraes & Paranjape, 2013). We conducted research with a goal to developing Information Technology (IT) Project Management (PM) methods to deal with these unsuspected challenges in BPR. This paper discusses one particular factor, which is how messy data may be a hidden failure factor that BPR projects typically may not detect during the planning phase. Our case study deals with SCM within the supply chains of two major urban hospitals, involving $2 million in minimum stocks for drug inventory. We used a mixed research method, gathering qualitative data on pharmaceutical supply-chain practices with interviews, observations, and internal documents analysis. We then performed statistical analysis on their inventory database systems, at all steps of the supply chain (e.g., prescriptions, orders, receptions, inventories, in-process, and delivery). Our project addresses the issue of the feasibility of the hospital’s Data Warehousing (DW) integration, especially at the stage of Extract, Transform, and Load (ETL). Our findings demonstrate that even the most rigorous planning for supply chain reengineering may rely on too many core assumptions built on inputs that could not have been detected until attempting to integrate data schema. We conclude with suggestions for modifying project-planning methods within supply chain management, particularly related to circumstances where messy data due to varying generations of information systems is a prevalent challenge. We prescribe an iterative and emergent Enterprise Architecture (EA) audit and verification method that may serve to guide reengineering project planning and execution. 2. LITERATURE REVIEW Process reengineering has typically relied on the introduction of more integrated and innovative IT in organizations. Systems integration, especially in the era of cloud computing and big data, remains a priority to develop more flexible and real-time EA. We briefly review the literature on reengineering and systems integration, especially in the case of DW, with particular attention to healthcare SCM. We identify to what extent prevailing project management methods allow for an early or later detection of messy data across systems. We identify the latest EA verification methods to assure the feasibility of process innovation. 2.1 BPR for SCM Planning a BPR project for improving SCM processes may present challenges. In the particular case of healthcare Keywords: Supply Chain Management (SCM), Business Process Management (BPM), Decision Support System (DSS), Extract Transform and Load (ETL), Data Models, Verification of Information System. Département des sciences administratives Université du Québec en Outaouais Gatineau, Québec, Canada (e-mail: veronique.nabelsi@uqo.ca, stephane.gagnon@uqo.ca) Abstract: This paper discusses how messy data may be a hidden failure factor that Business Process Reengineering (BPR) projects typically cannot detect during the planning phase. Our case study deals with Supply Chain Management (SCM) within two major urban hospitals, involving $2 million in minimum stocks for drug inventory. Our project addresses the feasibility of the hospital’s data warehousing integration, especially at the stage of Extract, Transform, and Load (ETL). We conclude with a proposed system architecture audit and verification method that may serve to guide reengineering project planning and execution. Detecting Constraints in Supply Chain Reengineering Projects: Case Study of Data and Process Integration in a Hospital Pharmacy Véronique Nabelsi and Stéphane Gagnon