Search

As interconnects dominate circuit performance in modern FPGAs, placement becomes a crucial stage for timing closure. Traditional FPGA placers seldom consider the timing constraints, and thus may lead to illegal routing solutions. In this paper, we present an incremental timing-driven placement flow for advanced FPGAs. First, a timing-based global placement strategy is designed to guide heterogeneous blocks to desired locations with satisfied timing constraints. Then, a timing-aware packing algorithm is developed to mitigate the design complexity while improving the timing results. Finally, we propose a critical path-based optimization method to generate optimized layout without timing violations. We evaluate our algorithm based on industrial circuits using an advanced FPGA device. The experimental results show that our placer achieves an 5.1% improvement in worst slack, and produces placements that require 16.7% less time to route when compared with the leading commercial tool Xilinx Vivado.

Human resources have played an important role in the production systems of manufacturing enterprises. At the same time, human resource allocation, as a scheduling problem, has attracted more and more attention from the industry and academia, with the increasing complexity of technology and the rising cost of workforce. However, the existing scheduling methods fail to fully consider the relationship between tasks and employees. In this paper we consider a single workgroup scheduling problem with the assumption of human resources variability. Jobs must be assigned to employees, where the number of processing personnel and the processing time are variable for one job. We model the problem as a combinatorial optimization problem and the objective considered is to minimize the maximum finish time which is called makespan. This problem is transformed into a two-dimension rectangle strip packing problem and we propose a hybrid optimization algorithm which combines scheduling algorithm with packing algorithm. Finally, different size cases are adopted to test the model. Results show the proposed strategy and algorithm are suitable for solving single workgroup scheduling very well.

Field programmable gate array (FPGA) packing and placement without routability consideration could lead to unroutable results for high-utilization designs. Conventional FPGA packing and placement approaches are shown to have severe difficulties to yield good routability. In this paper, we propose an FPGA packing and placement engine called UTPlaceF that simultaneously optimizes wirelength and routability. A novel physical and congestion aware packing algorithm and a hierarchical detailed placement technique are proposed. UTPlaceF outperforms state-of-the-art FPGA placers simultaneously in runtime and solution quality on International Symposium on Physical Design (ISPD) 2016 benchmark suite. Compared with the top three winners of ISPD’16 FPGA placement contest, UTPlaceF can deliver 6.2%, 11.6%, and 29.1% better routed wirelength with shorter runtime.

The paper proposes a way to examine whether cervical cells qualify as normal or abnormal on the basis of phase profiles, and to subsequently perform local fractal descriptions of their thre...

A packing algorithm for spatial multiplexing of orbital angular momentum states in a single diffractive structure is presented. Simulation results show the possibility to multiplex more than 200 states.

In Cloud Manufacturing environment, massive 3D Printing services in various types provide users with the ability of mass customization. The large number of 3D printing tasks bring more challenges on printing task scheduling to improve 3D printers’ utilization and thus saving time and materials. This paper establishes the model of 3D printing process in different types and figures out the existence of auxiliary processes. Aiming at decreasing the ratio of auxiliary process time consumption, this paper develops an algorithm derived from the rectangle packing problem to pack printing tasks whose model size are relatively small into one task and print them all in one 3D printing process. Experiments show that the ratio of auxiliary process time consumption significantly reduced by this algorithm.

Aiming at billet stacking problem of the rolling mill, an IBF algorithm is used to assign the billet to the warehouse positions. The utilization of non-empty warehouse positions and shuffles are considered in the algorithm. Computational experiments demonstrate that the proposed algorithm is feasible and effective, competitive and able to find more high-quality solutions for problem instances than artificial method.

[EN] Modified control charts are used to monitor and control manufacturing processes which are considered to be six-sigma processes. The use of these charts is based on the idea that the cost of identifying and correcting special causes is much higher than the cost of off-target products. Therefore, the process mean is essentially acceptable as long as it is anywhere within the specification limits. These concepts are applied to the packaging process in multihead weighers. The weight of the packed product, seen as the quality characteristic to be monitored, must be as close to a specified target weight as possible and comply with applicable regulations. The packaging process was previously optimized and improved using a packaging strategy, which was evaluated through a proposed packing algorithm. In this way, a set of numerical experiments were conducted to examine the solutions generated, which were subsequently monitored., [ES] Los gráficos de control modificados se utilizan para el seguimiento y control de procesos de fabricación que son considerados como procesos seis-sigma. El uso de estos gráficos se basa en la idea de que el costo de identificar y corregir causas especiales de variación es mucho más alto que el costo de productos alejados de su valor nominal. Por lo tanto, la media del proceso es esencialmente aceptable siempre que esté dentro de los límites de especificación. Estos conceptos han sido aplicados al proceso de envasado en pesadoras multicabezal. El peso del producto envasado, visto como la característica de calidad a ser monitoreada, debe ser lo más cercano posible a un peso objetivo y cumplir con la normativa. El proceso ha sido previamente optimizado y mejorado mediante una estrategia de envasado, la cual es evaluada a través de un algoritmo de envasado. De esta manera, un conjunto de experimentos numéricos fueron realizados para examinar las soluciones generadas, las cuales son posteriormente monitoreadas.

This paper proposes a new packing algorithm for strips cutting method in steel plate cutting technology. It remarks the shapes by many horizontal lines. It firstly searches the best sequence and each shape’s angle for packing all shapes by genetic simulated annealing algorithm, and then a new heuristic algorithm which satisfies the practical need of strips cutting technology based on the bottom-left algorithm is used to complete the automatic layout of two dimensional irregular shapes. In the end, satisfactory results of optimal layout are obtained.

This paper on the fundamental structural properties of shell-side packed columns presents an investigative study using a novel packing algorithm with the aim of validation against experimental results. The novel contribution in the predictive approach employed is the digitalisation of particles, container and packing space to generate complex life-like stochastic packing structures of particulate systems at relatively high speed (when compared to traditional vector-based approaches) which are comparable with experimentally packed beds without need for further conversion of numerical data. A variety of packed column simulations, comprising more complex beds than often seen in the literature, namely shell-side beds with increasing complexity of internal structure and packed with spherical particles, have been undertaken and the resulting averaged numerical data compared like-for-like with experimental results with key trends reproduced. The overall aim of this work is to demonstrate the usefulness of the technique and associated computer code as an engineering aide for research, design and optimisation of the performance of packed bed systems of any geometric complexity in terms of packing structure, as well as to highlight possible areas for further development of the code.

A new multi-threaded memetic algorithm (genetic algorithm with local optimizations) for the ISO 3832 luggage packing standard is presented. Apart from producing high-quality results, the algorithm also exploits the highly parallel computer architectures that are becoming more common as well as packing more boxes than other state-of-the-art algorithms. We also provide the first comparison of state-of-the-art packing algorithms for a real world trunk geometry.

The bin packing problem considered in this paper has the character with items arriving in batches, such as the task dispatching process in the cloud. The problem has two packing objects, one of which is to minimize the number of used bins, and the other is to reduce the packing time. To meet the objects, a container packing algorithm is presented, which consists of rounds of packing. In one round the number of the necessary bins is first calculated, and those bins are opened in batch to pack items in parallel. This round will be iterated until there are no unpacked items. The proposed algorithm is compared with other classical bin packing algorithms through experiments, which have the same property of simplicity for implementation. It is shown that the novel algorithm not only saves the packing time, but also achieves better combined performance than others.

The work presented in this paper givesmore insight and deeper understanding of constituting modules of parking system. This paperinvestigates the optimized parking system through FPGA employing “logic forgathers” including multiple registers as their logic block. And we show that this algorithm is superior to an existing packing algorithm.

In this paper, we propose an anti-jamming resource allocation scheme based on reserve channel with priority (RCP) fit algorithm for multi-frequency, time division multiple access (MF-TDMA) technology in satellite networks. The RCP fit allocates the traffic of the same terminal on the same frequency channel. In the RCP fit, partial bandwidth noise jamming (PBNJ) can corrupt the traffic of the specific terminals in jamming channels, and results in a fairness problem. Therefore, we extend the RCP fit to frequency hopping RCP (FH-RCP) fit, which prevents PBNJ using two steps. The first step is control channel frequency hopping to prevent continued jamming of the control channel, which forces the terminals to log out. The second step is data traffic frequency hopping, which enables various kinds of traffic to hop frequencies. FH-RCP fit thus applies the FH technique in a media access control layer, especially in MF-TDMA. Simulation results show that the proposed scheme improves anti-jamming capability and increases fairness compared to the RCP fit under PBNJ.

The packing of boxes distributed to different customers from a central depot is under consideration in this study. The boxes are of various sizes and they arrive on a conveyor belt one at a time. The on-line situation is assumed, which means that at each stage we only know the layout of the previous boxes on the partially filled pallet and the size of the box to be placed next but we have no information about the forthcoming ones. Further it is not allowed to move boxes already placed. The objective is to produce efficient and stable loads by an on-line packing algorithm, which can be implemented in automated systems. The algorithms we develop are based on the well-known first-fit and best-fit principles and their efficiency has been tested empirically.

With the advent of the great challenge brought by increasing complexity of modern large circuit, a pressing and necessary problem, that is, improving the routability and timing performance, is proposed in front of us. A novel packing algorithm called repack based on enhanced packing attraction function is presented while at the same time an iterative CAD flow tool could provide decreased interconnection resources requirement by applying CLB depopulation at given routing channel width limitation and local congested situations. Experimental results show that, for non-iterative flow, compared to the T-VPack and iRAC, repack can achieve 6.4% and 8.1% improvement respectively in timing performance. However, for iterative flow, when compared to T-VPack, repack has 12.6% and 37.6% improvement in area and routing path width respectively. When compared to iRAC, repack has a 0.9% decrease in area, but it has an improvement of 16.2% in routing path width instead.

We present a packing-based approximation algorithm for the k-Set Cover problem. We introduce a new local search-based k-set packing heuristic, and call it Restricted k-Set Packing. We analyze its tight approximation ratio via a complicated combinatorial argument. Equipped with the Restricted k-Set Packing algorithm, our k-Set Cover algorithm is composed of the k-Set Packing heuristic [8] for k≥7, Restricted k-Set Packing for k=6,5,4 and the semi-local (2,1)-improvement [2] for 3-Set Cover. We show that our algorithm obtains a tight approximation ratio of $H_k-0.6402+\Theta(\frac{1}{k})$ , where Hk is the k-th harmonic number. For small k, our results are 1.8667 for k=6, 1.7333 for k=5 and 1.5208 for k=4. Our algorithm improves the currently best approximation ratio for the k-Set Cover problem of any k≥4.

The rapid growth of band requirements for data transmission in computer networks is commonly known. The growth of the network infrastructure in many countries is much slower than expected. Therefore this paper presents some introductory research towards effective and inexpensive algorithmic ways for improving the efficiency of data transmission in all network layers. In this research human behavior is also taken into account.

Describes how an LP‐based one‐dimensional packing algorithm can be used to minimise costs in three different multiple‐vehicle shipment planning problems. The three applications considered are: (1) shipping industrial trucks by container when weight is the dominant constraint; (2) shipping automotive parts in racks by truck when volume is the dominant constraint; and (3) shipping large rolls of paper in railcars of varying size with incentive rates when volume is the dominant constraint.

This paper addresses the 3D packing problem of automatically determining positions in one or more containers or pallets for loading box-shaped goods and establishing a loading plan designed to achieve maximum loading capacity under various container- and cargo-related restrictions. In this paper, an algorithm for developing an efficient loading plan, based on an iterative local search method using a packing algorithm, will be introduced.

In this paper, algorithms which could improve the transmission band for computer networks has been presented. The paper presents the idea and implementation of a mobile agent which could the improve transmission band in a computer network, especially for HTTP files. As a result, the number of transmitted packets (frames) has been reduced several times and transmission efficiency has been improved.

The single-sequence (SS) is simply a sequence of integers 1, 2, 3, ..., n. But it leads a unique set of ABLR-relations (above, below, left-of, right-of) that hold among n objects on a plane. The direct relation set (DRS) is used to represent the direct ABLR-relations among rooms. It can be reused for any packing under the same floorplan. In this paper we analyze the relation between DRS and SS, and propose a general packing algorithm based on SS. It can update each room's coordinates on the fly. As a feature of SS, this algorithm covers all possible placements, if our objective is in bounding-box area minimization. Experimental results showed that the proposed algorithm is pretty fast and efficient compared with the performance attained by conventional algorithms.

Persson and Persson describe operating room scheduling wherein patients receive care within days [1]. The paper is similar to our work published in 2000 [2]. We speculate that the similarity arises from the authors’ assumptions being reasonable and realistic, as were ours. The insensitivity of results to the “on line” bin packing algorithm was published the preceding year (1999) [3], as was the requirement from the practical requirement from the integer nature of the problem to use at least a four-week period [4]. Persson and Persson are modeling the efficiency of use of OR time and yet lack the advances made in the >100 papers using those techniques since publication of Strum et al.’s 1999 paper [5]. Our 2000 paper [2] combined the research published independently in 1999. The paper from 2000 has 31 citations so reasonably could have been found by literature review. The two papers from 1999, on which the 2000 paper was based, have 42 and 77 citations, respectively.

Motivated by potential applications to computer storage allocation, we generalize the classical one-dimensional bin packing model to include dynamic arrivals and departures of items over time. Within this setting, we prove close upper and lower bounds on the worst-case performance of the commonly used First Fit packing algorithm, and, using adversary-type arguments, we show that no on-line packing algorithm can satisfy a substantially better performance bound than that for First Fit.

In telecommunications network planning, the problem of multiplexing low speed lines onto high speed transmission channels provides opportunities for attractive economic solutions. In many cases, the overall costs for transmission channels runs into hundreds of thousands of dollars per year. Thus, the problem of finding the minimum number of transmission channels to support low speed lines might be very important. This paper describes an exact packing algorithm (OPTPACK) for determining the packing of low speed lines onto high speed channels in such a way that the number of required channels is minimized.