Laura Barbulescu

• Carnegie Mellon University

In this paper, we consider the problem of feasibly in-tegrating new pick-up and delivery requests into exist-ing vehicle itineraries in a dynamic, dial-a-ride problem (DARP) setting. Generalizing from previous work in oversubscribed task scheduling, we define a controlled iterative repair search procedure for finding an alterna-tive set of vehicle...

Oversubscribed scheduling problems have been approached using both direct representations of the solution space and indirect, permutation-based representa- tions (coupled with a schedule builder to produce the corresponding schedule). In some problem contexts, permutation-based search methods have been shown to outperform schedule-space search meth...

In this paper, we describe the application of a multi-agent framework for collaborative scheduling to a disaster response coordination problem. The target problem is a field exer-cise mockup of a natural disaster, where a team of human agents must rely on their respective automated scheduling agents to coordinate and accomplish various infra-struct...

We consider the problem of coordinating a team of agents engaged in executing a set of inter-dependent, geographically dispersed tasks in an oversubscribed and uncertain environment. In such domains, where there are sequence-dependent setup activities (e.g., travel), we argue that there is inherent leverage to having agents maintain advance schedul...

Resource based scheduling using permutation based representations is reviewed. Permutation based representations are used in conjunction with genetic algorithms and local search algorithms for solving three very different scheduling problems. First, the Coors warehouse scheduling problem involves finding a permutation of customer orders that minimi...

In recent years, planning and scheduling research has paid increasing attention to problems that involve resource over- subscription, where cumulative demand for resources out- strips their availability and some subset of goals or tasks must be excluded. Two basic classes of techniques to solve oversubscribed scheduling problems have emerged: searc...

When organizing a scheduling competition one important is- sue is identifying the test problems for which the participating teams should demonstrate the performance of their approach. Ideally, the results of a competitive evaluation of different ap- proaches should be useful to both scheduling researchers and practitioners. To enable this, we belie...

The Air Force Satellite Control Network (AFSCN) coordinates communications to more than 100 satellites via nine ground stations positioned around the globe. Customers request an antenna at a ground station for a specific time window along with possible alternative slots. Typically, 500 requests per day result in more than 100 conflicts, which are r...

The best performing algorithms for a particular oversubscribed scheduling application, Air Force Satellite Control Network (AFSCN) scheduling, appear to have little in common. Yet, through careful experimentation and modeling of performance in real problem instances, we can relate characteristics of the best algorithms to characteristics of the app...

In this paper, we describe an incremental scheduling framework designed to support joint management of inter-dependent schedules by multiple executing agents. We assume an uncertain execution environment and a distributed representation of the overall problem and schedule such that no single agent has a complete view. Hence as unexpected execution...

This paper presents new results showing that a very simple stochastic hill climbing algorithm is as good or better than more complex metaheuristic methods for solving an oversubscribed scheduling problem: schedul- ing communication contacts on the Air Force Satellite Control Network (AFSCN). The empirical results also suggest that the best neighbor...

Representations are formalized as encodings that map the search space to the vertex set of a graph. We define the notion of bit equivalent encodings and show that for such encodings the corresponding Walsh coefficients are also conserved. We focus on Gray codes as particular types of encoding and present a review of properties related to the use of...

Oversubscribed scheduling problems require removing tasks when enough resources are not available. Prior AI approaches have mostly been constructive or repair-based heuristic search. In contrast, we have found a genetic algorithm (GA) to be the best approach to the overconstrained problem of Air Force Satellite Control Network scheduling. We presen...

Oversubscribed scheduling problems require removing or partially satisfying tasks when enough resources are not available. For a particular oversubscribed problem, Air Force Satellite Control Network scheduling, we find that the best approaches make long leaps in the search space. We find this is in part due to large plateaus in the search space. A...

We present the first coupled formal and empirical analysis of the Satellite Range Scheduling application. We structure our study as a progression; we start by studying a simplified version of the problem in which only one resource is present. We show that the simplified version of the problem is equivalent to a well-known machine scheduling problem...

Benchmarks and test suites are widely used to evaluate optimization and learning systems. The advantage is that these test problems provide an objective means of comparing systems. The potential disadvantage is that systems can become overfitted to work well on benchmarks and therefore that good performance on benchmarks does not generalize to real...

Three algorithms are tested on the satellite range scheduling problem, using data from the U.S. Air Force Satellite Control Network; a simple heuristic, as well as local search methods, are compared against a genetic algorithm on old benchmark problems as well as problems produced by a generator we recently developed. The simple heuristic works wel...

The use of random test problems to evaluate algorithm performance raises an important, and generally unanswered, question: Are the results generalizable to more realistic problems? Researchers generally assume that algorithms with superior performance on difficult, random test problems will also perform well on more realistic, structured problems....

Three algorithms are tested on the satellite range scheduling problem, using data from the U.S. Air Force Satellite Control Network; a simple heuristic, as well as local search methods, are compared against a genetic algorithm on old benchmark problems as well as problems produced by a generator we recently developed. The simple heuristic works wel...

A proof is presented that shows how the neighborhood structure that is induced under a Gray Code representation repeats under shifting. Convergence proofs are also presented for steepest ascent using a local search bit climber and a Gray code representation: for unimodal 1-D functions and multimodal functions that are separable and unimodal in each...

We compare several techniques for scheduling shipment of customer orders for the Coors Brewing warehouse and production line. The goal is to minimize time at dock for trucks and railcars while also minimizing inventory. The techniques include a genetic algorithm, local search operators, heuristic rules, systematic search and hybrid approaches. Init...

Test suites for many domains often fail to model features present in real-world problems. For the permutation flow-shop sequencing problem (PFSP), the most popular test suite consists of problems whose features are generated from a single uniform random distribution. Synthetic generation of problems with characteristics present in real-world proble...

Local search algorithms often get trapped in local optima. Algorithms such as tabu search and simulated annealing 'escape' local optima by accepting nonimproving moves. Another possibility is to dynamically change between representations; a local optimum under one representation may not be a local optimum under another. Shifting is a mechanism whic...

Many algorithms leverage structure to solve problems. In flowshop scheduling, we study two types of structure. First, for each job, we generate processing times that are similar across all machines (job-correlated problems). Then, for each machine, we generate processing times that are similar for all the jobs (machine-correlated problems). For the...

We compare several techniques for scheduling shipment of customer orders for the Coors Brewing warehouse and production line. The goal is to minimize time at dock for trucks and railcars while also minimizing inventory. The techniques include a genetic algorithm, local search operators, heuristic rules, systematic search and hybrid approaches. Init...

Both direct schedule representations as well as indirect permutation-based representations in con-junction with schedule builders are successfully employed in oversubscribed scheduling research. Recent work has indicated that in some domains, searching the space of permutations as opposed to the schedule space itself can be more productive in maxim...

Scheduling access to Air Force satellites is an over-constrained problem. Over 500 requests for access are received for each day's operation; after manual scheduling, over 100 connicts typically remain in the schedule. We are developing a prototype scheduling system that will assist human schedulers in this problem. Automated scheduling promises to...