Wen-Jing Hsu

Scheduling and routing algorithms for AGVs: A survey

Automated guided vehicles (AGVs) are now becoming popular in automated materials handling systems, flexible manufacturing systems and even container handling applications. In the past few decades, much research has been devoted to the technology of AGV systems and rapid progress has been witnessed. As one of the enabling technologies, scheduling and routing of AGVs have attracted considerable attention. Many algorithms for the scheduling and routing of AGVs have been proposed. However, most of the existing results are applicable to systems with a small number of AGVs, offering a low degree of concurrency. With a drastically increased number of AGVs in recent applications (e.g. in the order of a hundred in a container handling system), efficient algorithms are needed to resolve the increased contention of resources (e.g. path, loading and unloading buffers) among AGVs. This survey paper first gives an account of the emergence of the problem of AGV scheduling and routing. It then differentiates it from several related problems and classifies major existing algorithms for the problem. Finally, the paper points out fertile areas for future study of AGV scheduling and routing.

Fibonacci cubes-a new interconnection Topology

A novel interconnection topology called the Fibonacci cube is shown to possess attractive recurrent structures in spite of its asymmetric and relatively sparse interconnections. Since it can be embedded as a subgraph in the Boolean cube (hypercube) and it is also a supergraph of other structures, the Fibonacci cube may find applications in fault-tolerant computing. For a graph with N nodes, the diameter, the edge connectivity, and the node connectivity of the Fibonacci cube are in the logarithmic order of N. It is also shown that common system communication primitives can be implemented efficiently. >

Travel time analysis of a new automated storage and retrieval system

In conventional automated storage and retrieval systems (AS/RS), stacker cranes are used to access (store or retrieve loads into/from) the storage cells. The stacker cranes can travel simultaneously in the vertical and horizontal directions. However, because the combined motions generally require heavy machineries, the stacker cranes are inadequate for extra heavy loads such as sea container cargo. For such applications, we present a new kind of storage/retrieval (S/R) mechanism, designed with input from AS/RS manufacturers. Unlike stacker cranes, the new S/R mechanism has one vertical platform and N horizontal platforms to serve N tiers of an AS/RS rack. The vertical platform provides the vertical link among different tiers of the AS/RS rack, whereas the horizontal platforms access the storage cells on a given tier. The vertical platform and the horizontal platforms may move independently and concurrently; and the separation of the mechanisms for vertical/horizontal movements also makes the platforms lighter and hence they can operate at a higher speed than the conventional design. We then present a travel-time model under the stay dwell point policy, i.e. the platforms remain where they are after completing a storage/retrieval operation. The model is validated by computer simulations. The results show that our analytical model is reliable for the design and analysis of the new kind of AS/RS. We also present guidelines for the optimal design of a rectangular-in-time AS/RS rack with the new S/R mechanism.

The exchanged hypercube

This paper presents the exchanged hypercube, a new interconnection network obtained by systematically removing links from a binary hypercube. It maintains several desirable properties of the binary hypercube yet with reduced interconnection complexity. We also introduce the extended binomial tree, a spanning tree of the exchanged hypercube that preserves many desirable properties of the original binomial tree. A fault-tolerant routing strategy is also proposed for the exchanged hypercube.

Load balancing for conservative simulation on shared memory multiprocessor systems

Load balancing is a crucial factor in achieving good performance for parallel discrete event simulations. We present a load balancing scheme that combines both static partitioning and dynamic load balancing. The static partitioning scheme maps simulation objects to logical processes before simulation starts while the dynamic load balancing scheme attempts to balance the load during runtime. The static scheme involves two steps. First, the simulation objects that contribute to small lookahead are merged together by using a merging algorithm. Then a partitioning algorithm is applied. The merging is needed to ensure a consistent performance for our dynamic scheme. Our dynamic scheme is tailor-made for an asynchronous simulation protocol that does not rely on null messages. The performance study on a supply chain simulation shows that the partitioning algorithm and dynamic load balancing are important in achieving good performance.

How network topology affects dynamic loading balancing

Previous research has proposed several different load-balancing strategies and measured their performances on either a distributed system or a multiprocessor network of specific topology. The authors broadly classify all load-balancing strategies as being either static or dynamic. For certain applications, dynamic load balancing is preferable, because then the problems variable behavior more closely matches available computational resources. The authors address the performance of five dynamic load-balancing strategies: the Gradient Model strategy, the Sender-Initiated and Receiver-Initiated strategies, the Central Job Dispatcher strategy, and the Prediction-based strategy. The authors use a trace-driven simulation approach, collecting job traces from a production-distributed computer system and using them to simulate a loosely coupled multiprocessor network. This simulator enables performance comparisons across a range of network topologies, including a 2D-mesh, a 4D-hypercube, a linear array, and a composite Fibonacci cube.

Dynamic yard crane dispatching in container terminals with predicted vehicle arrival information

The performance of a container terminal depends on many aspects of operations. This paper focuses on the optimal sequencing of a yard crane (or YC for short) for serving a fleet of vehicles for delivery and pickup jobs. The objective is to minimize the average vehicle waiting time. While heuristic algorithms could not guarantee an optimal solution, a conventional mathematical formulation such as mixed integer program would require too much computing time. We present two new algorithms to efficiently compute YC dispatching sequences that are provably optimal within the planning window. The first algorithm is based on the well-known A^* search along with an admissible heuristics. We also incorporate this heuristics into a second backtracking algorithm which uses a prioritized search order to accelerate the computation. Experimental results show that both new algorithms perform very well for realistic YC jobs. Specifically, both are able to find within seconds optimal solutions for heavy workload scenarios with over 2.4x10^1^8 possible dispatching sequences. Moreover, even when the vehicle arrival times are not accurately forecasted, the new algorithms are still robust enough to produce optimal or near-optimal sequences, and they consistently outperform all the other algorithms evaluated.

Survey of Languages and Runtime Libraries for Parallel Discrete-Event Simulation

To develop a parallel discrete-event simulation from scratch requires in-depth knowledge of the mapping process from the physical model to the simulation model, and also a substantial effort in coping with the numerous issues concerning the underlying synchronization protocols in use. Languages and libraries could reduce the devel opment effort significantly by providing the user with a pre-built parallel simulation kernel as well as application development tools. This pa per contains a survey of the existing languages and libraries for parallel discrete-event simula tion. It is divided into two major sections: one on the languages, the other on the libraries. The discussions are mainly focused on the following aspects: user model, programming framework and language features, library API, protocols, and system support and environment. The re ported performances of some packages are also summarized.

Efficient Adaptive Scheduling of Multiprocessors with Stable Parallelism Feedback

With proliferation of multicore computers and multiprocessor systems, an imminent challenge is to efficiently schedule parallel applications on these resources. In contrast to conventional static scheduling, adaptive schedulers that dynamically allocate processors to jobs possess good potential for improving processor utilization and speeding up jobs execution. In this paper, we focus on adaptive scheduling of malleable jobs with periodic processor reallocations based on parallelism feedback of the jobs and allocation policy of the system. We present an efficient adaptive scheduler Acdeq that provides parallelism feedback using an adaptive controller A-Control and allocates processors based on the well-known Dynamic Equipartitioning algorithm (Deq). Compared to A-Greedy, an existing adaptive scheduler that experiences feedback instability thus incurs unnecessary scheduling overheads, we show that A-Control achieves much more stable feedback among other desirable control-theoretic properties. Furthermore, we analyze algorithmically the performances of Acdeq in terms of its response time and processor waste for an individual job as well as makespan and total response time for a set of jobs. To the best of our knowledge, Acdeq is the first multiprocessor scheduling algorithm that offers both control-theoretic and algorithmic guarantees. We further evaluate Acdeq via simulations by using Downeys parallel job model augmented with internal parallelism variations. The results confirm its improved performances over Agdeq, and they show that Acdeq excels especially when the scheduling overhead becomes high.

Yard crane dispatching based on real time data driven simulation for container terminals

This paper studies the problem of real time yard crane dispatching in container terminals. Many technologies, including transponders, RFID and GPS have been used in the container terminal setting for real-time tracking of terminal equipment. A judicious integration of real-time data into the yard crane management system will allow better utilization of terminal resources to improve overall terminal productivity. We propose a yard crane dispatching algorithm based on real time data driven simulation to solve the problem of yard crane job sequencing to minimize average vehicle waiting time. The algorithm will produce optimal operation sequence for each planning window. Several policies to select jobs to form the planning window are also proposed. Our simulation results show that dispatching yard crane based on real time data driven simulation is of great value in improving yard crane performance in 3 scenarios with different vehicle arriving patterns and our results are 10% worse off a loosely estimated overall optimal performance result.