Wade Cook

Project portfolio management (PPM) is an important area of interest in many organizations. There is a wide literature on each of many different aspects of PPM. The central purpose of the current paper is to focus on a specific sub-area of PPM, namely the project portfolio selection (PPS) problem. Specifically, we develop a new methodology that will aid management in choosing from a set of candidate project proposals, a subset of those project proposals that align with strategic objectives of the organization. Research methodology is based on the data envelopment analysis (DEA) construct to compare a set of decision making units (such as proposed projects) to arrive at an efficiency score for each member of this competing set, derive the best performers, generate an efficiency frontier and quantify inefficiency in the non-best performers. While DEA has been applied in numerous settings, the unique feature of the project portfolio application is the presence of two sets of data, namely pre-implementation "estimates", and post-implementation "actuals". Our methodology is unique in that it uses the idea of dual DEA frontiers based on such before and after data for a set of past projects. Dual frontier concept makes not only an important practical contribution to the PPS literature, but as well it opens new directions and provides an innovative advancement in the DEA literature. The requisite data is not publicly available. Therefore, we develop a general methodology to illustrate our technique.

Wade D. Cook Schulich School of Business, York University, Toronto, Ontario M3J 1P3, Canada, wcook@schulich.yorku.ca Boaz Golany Faculty of Industrial Engineering and Management, Technion?Israel Institute of Technology, Haifa 32000, Israel, golany@ie. ...

ABSTRACT We introduce a branch-and-cut algorithm to aggregate published journal rankings based on subsets of the accounting literature in order to create a consensus ranking. The aggregate ranking allows specialist and regional journals, which may only be ranked in a limited number of studies, to be placed with respect to each other and with respect to the generalist journals that are usually included in ranking studies. The approach we develop is a significant advance over ad hoc approaches to aggregating journal rankings that have appeared in the literature and may provide a theoretically sound and replicable basis for further exploration of the concept of journal quality and the stability of journal rankings over time and ranking methods.Compilation de listes incomplètes de revues: une application aux revues comptables spécialiséesRésuméLes auteurs ont recours à un algorithme de séparation et coupes pour agréger les classements de revues, à partir de sous-ensembles de publications comptables, en vue de créer une classification consensuelle. La classification agrégée permet d’ordonner les revues spécialisées et régionales, qui ne sont classées que dans un nombre limité d’études, les unes par rapport aux autres ainsi que par rapport aux revues non spécialisées qui figurent habituellement dans les études de classification. La méthode élaborée par les auteurs est sensiblement supérieure aux méthodes ponctuelles de compilation des classements de revues proposées dans la documentation et peut servir debase théorique solide et renouvelable pour une exploration plus approfondie des notions de qualité des revues et de stabilité des classements dans le temps et selon les méthodes de classification.