Articles scientifiques

Analyzing Degree of Parallelism for Concurrent Timed Workflow Processes With Shared Resources

Yanhua DU, Li WANG, X. LI

IEEE Transactions on Engineering Management

février 2017, vol. 64, n°1, pp.42 - 56

Départements : Informations Systems and Operations Management, GREGHEC (CNRS)

Mots clés : Business, Servers, Computational modeling, Uncertainty, Analytical models, Processor scheduling, workflow management, Business process management, degree of parallelism, Petri net (PN), timed workflow net (TWF-Net)

Degree of parallelism is an important factor in workflow process management, because it is useful to accurately estimate the server costs and schedule severs in workflow processes. However, existing methods that are developed to compute degree of parallelism neglect to consider activities with uncertain execution time. In addition, these methods are limited in dealing with the situation where activities in multiple concurrent workflow processes use shared resources. To address the limitations, we propose a new approach to analyzing degree of parallelism for concurrent workflow processes with shared resources. Superior over the existing methods, our approach can compute degree of parallelism for multiple concurrent workflow processes that have activities with uncertain execution time and shared resources. Expectation degree of parallelism is useful to estimate the server costs of the workflow processes, and maximum degree of parallelism can guide managers to allocate severs or virtual machines based on the business requirement. We demonstrate the application of the approach and evaluate the effectiveness in a real-world business scenario.

Pricing and Capacity Allocation for Shared Services


Manufacturing & Service Operations Management

printemps 2017, vol. 19, n°2, pp.230-245

Départements : Informations Systems and Operations Management

Mots clés : customer mix; customer interaction; price discrimination; capacity allocation; shared services

We study the pricing and capacity allocation problem of a service provider who serves two distinct customer classes. Customers in each class are inherently heterogeneous in their willingness to pay for service, but their utilities are also affected by the presence of other customers in the system. Specifically, customer utilities depend on how many customers are in the system at the time of service as well as who these other customers are. We find that if the service provider can price discriminate between customer classes, pricing out a class, i.e., operating an exclusive system, can sometimes be optimal and depends only on classes’ perceptions of each other. If the provider must charge a single price, an exclusive system is even more likely. We extend our analysis to a service provider who can prevent class interaction by allocating separate capacity segments to the two customer classes. Under price discrimination, allocating capacity is optimal if the “net appreciation” between classes, as defined in the paper, is negative. However, under a single-price policy, allocating capacity can be optimal even if this net appreciation is positive. We describe in detail how the nature of asymmetry in classes’ perception of each other determines the optimal strategy

Strategic Investment in Renewable Energy Sources: The Effect of Supply Intermittency


Manufacturing & Service Operations Management

Summer 2017, vol. 19, n°3, pp.489-507

Départements : Informations Systems and Operations Management, GREGHEC (CNRS)

Mots clés : Electricity Generation, Renewables, Intermittency, Capacity Planning and Investment, Incentives and Contracting

To analyze incentives for investing in the capacity to generate renewable electricity, we model the trade-off between renewable (e.g. wind) and nonrenewable (e.g. natural gas) technology. Renewable technology has a higher investment cost and yields only an intermittent supply of electricity; nonrenewable technology is reliable and has lower investment cost but entails both fuel expenditures and carbon emission costs. With reference to existing electricity markets, we model several interrelated contexts - the vertically integrated electricity supplier, market competition, and partial market competition with long-term fixed-price contracts for renewable electricity - and examine the effect of carbon taxes on the cost and share of wind capacity in an energy portfolio. We find that the intermittency of renewable technologies drives the effectiveness of carbon pricing mechanisms, which suggests that charging more for emissions could unexpectedly discourage investment in renewables. We also show that market liberalization may reduce investment in renewable capacity while increasing the overall system's cost and emissions. Fixed-price contracts with renewable generators can mitigate these detrimental effects, but not without possibly creating other problems. In short: actions to reduce the intermittency of renewable sources may be more effective than carbon taxes alone at promoting investment in renewable generation capacity

Topological network design of closed finite capacity supply chain networks


Journal of Manufacturing Systems

octobre 2017, vol. 45, pp.70-81

Départements : Informations Systems and Operations Management

Mots clés : Closed;Finite networks;M/M/1/K;M/G/c/c;M/G/∞ queues

In this paper, we examine the layout, location, and general topological arrangement of queues in a closed finite queueing network environment for supply chains. Since our focus is on manufacturing environments, then maximizing throughput is a worthy performance measure objective. We are given a network topology G(V, E) with a finite set of nodes and edges and we wish to assign the queues to the nodes V ∈ G such a way that the maximum throughput is achieved. This is a nonlinear continuous optimization problem with implicit integer variables so the problem is at least NP-Hard. We also examine the impacts of Additive Manufacturing (AM) on the throughput of the supply chain. Decentralization of the supply chain topology as evidence by the increased dispersal of nodes within the topology tends to increase the throughput of the system, so the AM leaf nodes can have a measurable and significant impact on SCM throughput.


Informations Systems and Operations Management

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


Joseph NEHME

Informations Systems and Operations Management

Voir le CV