Dealing with supply uncertainty in liquid gas bulk supply chain

Informations Systems and Operations Management

Intervenant : Nicoleta Neagu
Operations Research Expert , Air Liquide

29 novembre 2013 - HEC - Campus Jouy en Josas - Bâtiment S - Salle S118 - De 14h00 à 15h00

A main challenge for supply chain management and operational decisions is data uncertainty due, for example, to supply disruptions, sudden unavailable resources, or unusual demand. Air Liquide is particularly concerned with uncertainty in the supply of bulk gas due to plant outages which generate significant losses and impact quality of service. Leveraging operations research techniques in robust and stochastic optimization, we propose methodologies for handling supply uncertainty due to plant outages at both the strategic and the operational levels. In this talk, I will first consider the strategic problem of optimizing customers' allocation coupled with production planning. The main objective is to minimize a total estimated cost which includes production, distribution, and contractual costs over a set of plant outages scenarios. Second, I will address the inventory routing problem for bulk gas distribution under supply uncertainty. I will present the general methodology we proposed for generating, classifying, and selecting ‘robust’ solutions which, by definition, are less impacted when plant outages occur. Finally, I will present some of the results we obtained when applying this methodology to Air Liquide's bulk supply chain.

Nicoleta Neagu received her MS in Computer Science from the University of Bucharest in 1998 and her Ph.D. in Artificial Intelligence from the Ecole Polytechnique Fédérale de Lausanne (EPFL) in 2003. From 2004 to 2006 she was a researcher at Whitestein Technology AG in Zurich. From 2006 to 2008, she was a postdoctoral associate at the Operations Research Center at the Massachusetts Institute of Technology (MIT), Cambridge, USA, where she worked on logistics problems in collaboration with United Parcel Service (UPS). Since 2008, Dr. Neagu has been with the Applied Mathematics group at the Claude Delorme Research Center of Air Liquide where she is now an Operations Research Expert.

Multisourcing newsvendor problem with a service level constraint: a distributionnally robust solution

Informations Systems and Operations Management

Intervenant : Jean-Philippe VIAL
Professeur , HEC Genève - Université de Genève

7 novembre 2013 - HEC - Campus Jouy en Josas - Bâtiment S - Salle S120 - De 14h00 à 15h30

The multi-sourcing Newsvendor problem is highly relevant as well as in the context of a reseller or a producer, but it poses challenging issues. In this problem, the decision maker needs to cover an uncertain demand by ordering from multiple suppliers. Each supplier faces a random yield factor with a general probability distribution.
The problem amounts to selecting how much to order from each supplier, so as to minimize expected procurement cost. Each supplier is characterized by given costs and a distribution of his random yield factor.
Determining the optimal ordering quantities is prohibitively difficult. We consider a maximin expected cost model, where the objective function is the worst-case expected cost over a set of probability distributions with given mean and covariance information. We derive the corresponding exact and approximate optimal ordering strategies. The optimal orders are identified using a tractable conic programming approach.

Cultivating an Electronic Health Record

Informations Systems and Operations Management

Intervenant : Professor Deborah COMPEAU
Richard Ivey School of Business , University of Western Ontario

26 juin 2013 - Class Room 119 New Building S - De 10h30 à 12h00

The development of interoperable electronic health records (EHRs) has been a challenge for many national jurisdictions. Highly publicized failures in the UK and Canada, along with slow progress towards adoption in many regions, calls into question the prevailing mindset about how to successfully implement these important systems. EHRs need to be understood and managed, not as large systems, but rather as information infrastructures. Using this positioning and drawing on prior research that takes an information infrastructure perspective, this paper evaluates a particular EHR tool, and shows how a system can be effectively cultivated and produce benefits to the adopters, despite both system quality and service quality limitations.

Deborah (Debbie) Compeau is a Professor of Management Information Systems at the Ivey Business School. She holds both HBA and PhD degrees from Ivey.
On faculty at Ivey since 2000, she has served as Director of HBA Student Relations (2005-2007) and as Director of the PhD Program (2007-2011). She has taught in the HBA, MBA, EMBA and PhD programs, as well as on the Ivey Management Development Program. She has also taught workshops on case teaching and learning in the US, Germany, France and China.
Professor Compeau's research focuses on the interaction between people and information technologies (IT) in organizations. She studies issues related to IT learning and adoption, in order to understand one of the critical mechanisms by which value is created through IT. Her specific interests include end user training and learning, and the adoption and implementation of new IT. In particular she is interested in understanding what organizations can do to facilitate individual adoption of and learning about information and communication technologies. Recent research projects have focused on the adoption of IT in healthcare settings.
Her research has been published in leading journals including Information Systems Research and MIS Quarterly, and has been recognized (Lowry et al (2007) Assessing Leading Institutions, Faculty, and Articles in Premier Information Systems Research Journals, Communications of AIS, Volume 20, Article 16), as among those with the highest impact. Her research on computer self-efficacy has received more than 700 citations according to Web of Knowledge (>2400 citations on Google Scholar). She has served in editorial roles at MIS Quarterly and Information Systems Research, and as a member of one of SSHRC's grants adjudication committees.

Decomposing the Effect of Workload on Patient Outcomes

Informations Systems and Operations Management

Intervenant : Nicos SAVVA
Professeur Assistant en Science du Management et des Opérations , London Business School

7 juin 2013 - HEC - Campus Jouy en Josas - Bâtiment S - Salle S121 - De 10h30 à 12h00

Extant literature has documented that in-patient outcomes deteriorate at high workloads. In this paper, we present our work with the Delivery Unit (DU) of a large teaching hospital that investigates two distinct mechanisms through which workload affects outcomes: (i) the amount of resources dedicated to individual patient care, and (ii) the quality of execution given the resource provision. In the context of the DU there are two well defined pathways of care, natural and assisted deliveries, that vary in the level of resources required. To quantify outcomes we use two measures, one medical -- the occurrence of a severe perennial laceration during childbirth, and one operational -- the post-birth length of stay (LOS). To measure workload we use an exceptionally detailed 4-year-long patient and staffing database. To overcome problems associated with endogenous selection (i.e. patients are more likely to be given assistance because they are more likely to experience complications), we formulate and estimate a simultaneous recursive bivariate probit model with appropriate exclusion restrictions. We find that the chance of receiving a more resource intensive treatment is significantly reduced and outcomes (both in terms of propensity of lacerations and LOS) deteriorate as workload increases. Nevertheless, once we control for endogeneity, we find that the decrease in the propensity of administering resource intensive treatments does not cause a further deterioration in outcomes, suggesting that at low workload levels there is some overtreatment. We discuss our findings operational (and medical) implications.

Joint work with Michael Freeman (Cambridge) and Stefan Scholtes (Cambridge)

Embarrassing Exposure in Online Social Networks: Privacy Invasion and Relationship Bonding"

Informations Systems and Operations Management

Intervenant : Zhenhui Jack Jiang
Associate Professor in the Department of Information Systems and Director of Centre for Collaborative Media and Technologies , National University of Singapore

29 avril 2013 - Class Room 119 Building S - De 10h00 à 11h30

People will feel uneasy if private embarrassing information about them is exposed in online social networks. While individuals are generally prudent in controlling access to their embarrassing information, such control is fundamentally challenged when the embarrassing information is co-owned with others. Drawing on the social exchange framework as the overarching theory, this paper examines the effects of information dissemination and network mutuality on perceived privacy invasion as well as perceived relationship bonding and how these perceptions shape individuals’ behavioral responses to an embarrassing exposure (i.e., inaction, avoidance, and approach). The results of a laboratory experiment involving 109 subjects provide strong evidence that information dissemination and network mutuality jointly influence individuals’ perceived privacy invasion and perceived relationship bonding. In addition, while perceived privacy invasion encourages transactional avoidance, it discourages approach behavior. Further, whereas perceived relationship bonding impedes both transactional avoidance and interpersonal avoidance, it leads to approach behavior. Overall, this study contributes to the IS literature by deepening the understanding of individuals’ behavioral response to embarrassing exposures triggered by others’ revelation

He obtained his Ph.D. degree in management information systems from the University of British Columbia, Vancouver, Canada, and Master’s and Bachelor degrees from Tsinghua University, Beijing. His research interests include the design and evaluation of social media, electronic commerce, online social network, and information privacy. His work has been published in top information systems journals, such as MIS Quarterly, Information Systems Research, Journal of the Association for Information Systems, and Journal of Management Information Systems. Dr. Jiang is serving as the Associate Editor of MIS Quarterly, Journal of the Association for Information Systems, and IEEE Transactions on Engineering Management. He was previously on the editorial boards of AIS Transactions on Human-Computer Interaction and Journal of Database Management.


Informations Systems and Operations Management

Campus HEC Paris
1, rue de la Libération
78351 Jouy-en-Josas cedex


Christian VAN DELFT

Informations Systems and Operations Management (GREGHEC)

Voir le CV