Optimality of index policies for stochastic scheduling with switching penalties

Oyen, Mark P. Van; Pandelis, Dimitrios G.; Teneketzis, Demosthenis

doi:10.1017/s0021900200043825

Cited by 12 publications

(17 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Priority-based scheduling has been studied in the context of queueing theory [49,[81][82][83][84][85][86][87]. It has been stated, that "there are no undiscovered priority index sequencing rules for minimizing total delay costs" [82].…”

Section: Optimization Strategymentioning

confidence: 99%

See 1 more Smart Citation

Self-control of traffic lights and vehicle flows in urban road networks

2008

View full text Add to dashboard Cite

Abstract.Based on fluid-dynamic and many-particle (car-following) simulations of traffic flows in (urban) networks, we study the problem of coordinating incompatible traffic flows at intersections. Inspired by the observation of self-organized oscillations of pedestrian flows at bottlenecks [D. Helbing and P. Molnár, Phys. Rev. E 51 (1995) 4282-4286], we propose a self-organization approach to traffic light control. The problem can be treated as multi-agent problem with interactions between vehicles and traffic lights. Specifically, our approach assumes a priority-based control of traffic lights by the vehicle flows themselves, taking into account short-sighted anticipation of vehicle flows and platoons. The considered local interactions lead to emergent coordination patterns such as "green waves" and achieve an efficient, decentralized traffic light control. While the proposed self-control adapts flexibly to local flow conditions and often leads to non-cyclical switching patterns with changing service sequences of different traffic flows, an almost periodic service may evolve under certain conditions and suggests the existence of a spontaneous synchronization of traffic lights despite the varying delays due to variable vehicle queues and travel times. The self-organized traffic light control is based on an optimization and a stabilization rule, each of which performs poorly at high utilizations of the road network, while their proper combination reaches a superior performance. The result is a considerable reduction not only in the average travel times, but also of their variation. Similar control approaches could be applied to the coordination of logistic and production processes.

show abstract

Section: Optimization Strategymentioning

confidence: 99%

“…However, the considered prioritization strategies were restricted to functions of the current queue length, i.e. to the number of vehicles that have already been stopped [81,88]. In contrast, our anticipation model (see Sec.…”

Section: Optimization Strategymentioning

confidence: 99%

Self-control of traffic lights and vehicle flows in urban road networks

2008

View full text Add to dashboard Cite

show abstract

“…Switching jobs from one machine to another incurs a variety of setup costs (Duenyas and Van Oyen (1996), Karaesmen and Gupta (1997), Reiman and Wein (1998), ) and switching costs (Van Oyen and Teneketzis (1993), Koole (1997), Van Oyen et al (1992), Kolonko and Benzing (1985)). Unfortunately, Bank and Sundaram (1994) showed that in the presence of switching costs it is not possible to define an index for each arm such that the resulting strategy produces the maximized payoffs.…”

Section: Switching Costsmentioning

confidence: 99%

“…The objective is to minimize the expected costs of serving all the jobs. The special case where there is only one job in a family is completely characterized by Santos and Magazine (1985) in a deterministic model, and Van Oyen et al (1992) in a stochastic model. Let x n denote the number of identical jobs in family n. The index for family n is equal to…”

Section: Van Oyen and Pichitlamken (1999)mentioning

confidence: 99%

A survey on the bandit problem with switching costs

Jun

2004

De Economist

View full text Add to dashboard Cite

SummaryThe paper surveys the literature on the bandit problem, focusing on its recent development in the presence of switching costs. Switching costs between arms makes not only the Gittins index policy suboptimal, but also renders the search for the optimal policy computationally infeasible. This survey will first discuss the decomposability properties of the arms that make the Gittins index policy optimal, and show how these properties break down upon the introduction of costs on switching arms. Having established the failure of the simple index policy, the survey focus on the recent efforts to overcome the difficulty of finding the optimal policy in the bandit problem with switching costs: characterization of the optimal policy, exact derivation of the optimal policy in the restricted environments, and lastly approximation of optimal policy. The advantages and disadvantages of the above approaches are discussed.

show abstract

“…They partially characterized the optimal policy and developed some dynamic heuristics. Van Oyen et al [27] treated versions without arrivals but with heterogeneous service processes and either set-up costs or set-up times. They proved that an index policy is optimal and computed the optimal indices for cases with lump-sum set-up costs and those with set-up times.…”

Section: Introductionmentioning

confidence: 99%

Stochastic scheduling of parallel queues with set-up costs

Duenyas

Oyen

1995

Queueing Syst

Self Cite

View full text Add to dashboard Cite

We consider the problem of allocating a single server to a system of queues with Poisson arrivals. Each queue represents a class of jobs and possesses a holding cost rate, general service distribution, and a set-up cost. The objective is to minimize the expected cost due to the waiting of jobs and the switching of the server. A set-up cost is required to effect an instantaneous switch from one queue to another. We partially characterize an optimal policy and provide a simple heuristic scheduling policy. The heuristic's performance is evaluated in the cases of two and three queues by comparison with a numerically obtained optimal policy. Simulation results are provided to demonstrate the effectiveness of our heuristic over a wide range of problem instances with four queues.

show abstract

Optimality of index policies for stochastic scheduling with switching penalties

Cited by 12 publications

References 19 publications

Self-control of traffic lights and vehicle flows in urban road networks

Self-control of traffic lights and vehicle flows in urban road networks

A survey on the bandit problem with switching costs

Stochastic scheduling of parallel queues with set-up costs

Contact Info

Product

Resources

About