Search

We examine a service system that matches jobs and candidates with different ability levels. A candidate’s ability will affect the possibility of obtaining a high return on a job assignment. Based on the property that the jobs to be assigned to a candidate are substituted in an assignment problem, we propose a novel random parameter generating process to demonstrate that the expected reward increases with the ability level of a candidate at a diminishing rate. This observation is consistent with the classic result in process flexibility literature that a little flexibility (i.e., ability level) can capture the most benefit. Our observations can help a company decide the proper level of ability.

A high-accuracy unsupervised classification model is developed in a multimode self-interference microring resonator (SIMRR). For the SIMRR, there are many whispering gallery modes (WGMs) present. Each of these resonance modes supported by the SIMRR has a different response to different target parameters, so that the SIMRR-based sensor has the super capability to distinguish between multiple components. In the classification model, principal component analysis (PCA) is firstly used to reduce the dimensionality of this multimode sensing information from the SIMRR-based sensor. When the original higher-dimensional data points are projected onto the lower-dimensional data with only the first few principal components, they can be easily categorized into several different types by using density-based spatial clustering of application with noise (DBSCAN) algorithm. As an example, the unsupervised classification method is numerically validated based on a designed three-gas SIMRR-based sensor. The numerical results prove that the classification model can achieve an ultra high classification accuracy for the designed three-gas sensor with more than 60 dB in signal-to-noise ratio.

Due to the low detection accuracy of YOLOv3 target detection method, this paper proposes an improved target detection method of YOLOv3 remote sensing image. Firstly, the feature extraction network DarkNet53 is strengthened to improve the ability of feature extraction; Secondly, the original Leaky ReLU activation function is replaced by the Mish activation function, thus improving the generalization of the method in this paper; Finally, the Learning rate and BatchSize parameters are modified to prevent overfitting. The remote sensing image datasets of RSOD and TGRS-HRRSD are used in this paper. The Average Precision (AP) results of the method on the RSOD datasets in this paper show that the mAP value is 5.33 percent higher than that of the previous YOLOV3 method. The log average miss-rate (LAMR) results show that the LAMR value is 0.1100 lower than that of the previous YOLOV3 method. The mAP results of the method on the TGRS-HRRSD datasets show that the MAP value is 1.29 percent higher than that of the previous YOLOV3 method, and the LAMR results show that the LAMR value is 0.0338 lower than that of the previous YOLOv3 method.

Deep learning based inpainting methods have obtained promising performance for image restoration, however current image inpainting methods still tend to produce unreasonable structures and blurry textures when processing the damaged images with heavy corruptions. In this paper, we propose a new image inpainting method termed Global Context Modeling Network (GCM-Net). By capturing the global contextual information, GCM-Net can potentially improve the performance of recovering the missing region in the damaged images with irregular masks. To be specific, we first use four convolution layers to extract the shadow features. Then, we design a progressive multi-scale fusion block termed PMSFB to extract and fuse the multi-scale features for obtaining local features. Besides, a dense context extraction (DCE) module is also designed to aggregate the local features extracted by PMSFBs. To improve the information flow, a channel attention guided residual learning module is deployed in both the DCE and PMSFB, which can reweight the learned residual features and refine the extracted information. To capture more global contextual information and enhance the representation ability, a coordinate context attention (CCA) based module is also presented. Finally, the extracted features with rich information are decoded as the image inpainting result. Extensive results on the Paris Street View, Places2 and CelebA-HQ datasets demonstrate that our method can better recover the structures and textures, and deliver significant improvements, compared with some related inpainting methods.

Deep denoiser, the deep network for denoising, has been the focus of the recent development on image denoising. In the last few years, there is an increasing interest in developing unsupervised deep denoisers which only call unorganized noisy images without ground truth for training. Nevertheless, the performance of these unsupervised deep denoisers is not competitive to their supervised counterparts. Aiming at developing a more powerful unsupervised deep denoiser, this paper proposed a data augmentation technique, called recorrupted-to-recorrupted (R2R), to address the overfitting caused by the absence of truth images. For each noisy image, we showed that the cost function defined on the noisy/noisy image pairs constructed by the R2R method is statistically equivalent to its supervised counterpart defined on the noisy/truth image pairs. Extensive experiments showed that the proposed R2R method noticeably outperformed existing unsupervised deep denoisers, and is competitive to representative supervised deep denoisers.

Existing mechanical HVDC circuit breakers (CB) are capable of interrupting HVDC currents within several tens of milliseconds, but this is too slow to satisfy the requirements of a reliable HVDC grid. HVDC CB based on semiconductors can easily overcome the limitations of operating speed, but need large number of the series (parallel) connection of electronic switching devices. To overcome these shortcomings, this paper introduces a topology of hybrid CB; this paper first describes the principle of hybrid CB; then a control strategy of fault pre-treatment is also proposed, that is, the ‘pre-action and action (or recovery)’ of the switches, in order to improve the breaking speed of the hybrid CB and the equalising circuit of its solid-state switch has been designed. The proposed topology and control strategy do not produce loss in the normal operation of CB, and instantaneous trigger solid-state switch with quickly cutting short-circuit current during short-circuit faults; at last the simulation and experiment of the hybrid HVDC CB have also been done. The results turn out when a short circuit occurs, the hybrid HVDC CB can quickly break the short-circuit current; the buffer circuit can be provided for the release of the stored energy circuit.

Compressed sensing (CS) is a signal processing framework for effectively reconstructing signal from a small number of measurements obtained by linear projections of the signal. It is an ongoing challenge for the real-time image reconstruction of the computational imaging, including single pixel imaging based on CS. We built a block-based CS (BCS) image reconstruction framework via a deep learning network with smoothed projected Landweber (SPL). A fully connected network performs both BCS linear sensing and non-linear reconstruction stages, and SPL removes the blocking artifacts due to incorporate Wiener filtering into projected Landweber (PL) method at each iteration. The sensing matrix and nonlinear prediction operator are jointly optimized, and the smoothing filtering is coalesced into the PL framework for eliminating high-frequency oscillatory blocking artifact. Experimental results reveal that the optimized scheme outperforms the approach only based on deep neural network. The reconstruction quality can be improved while being only slightly slower, especially the gain of structural similarity is significantly better than peak signal-to-noise ratio, and the reconstruction image texture details are vivid and natural. At 10% sensing rate, the structural similarity maximum (minimum) gain reaches 0.098 (0.021). The proposed approach is not only far superior to other state-of-the-art CS algorithms in terms of reconstruction time and quality but also comparable with up-to-date deep learning methods.

The use of non-local self-similarity prior between image blocks can improve image reconstruction performance significantly. We propose a compressive sensing image reconstruction algorithm that combines bilateral total variation and nonlocal low-rank regularization to overcome over-smoothing and degradation of edge information which result from the prior reconstructed image. The proposed algorithm makes use of the preservation of image edge information by bilateral total variation operator to enhance the edge details of the reconstructed image. In addition, we use weighted nuclear norm regularization as a low-rank constraint for similar blocks of the image. To solve this convex optimization problem, the Alternating Direction Method of Multipliers (ADMM) is employed to optimize and iterate the algorithm model effectively. Experimental results show that the proposed algorithm can obtain better image reconstruction quality than conventional algorithms with using total variation regularization or considering the nonlocal structure of the image only. At 10% sampling rate, the peak signal-to-noise ratio gain is up to 2.39 dB in noiseless measurements compared with Nonlocal Low-rank Regularization (NLR-CS). Reconstructed image comparison shows that the proposed algorithm retains more high frequency components. In noisy measurements, the proposed algorithm is robust to noise and the reconstructed image retains more detail information.

We propose two recommendation algorithms for e-commerce with supply limits, a scenario that has not been intensively studied in the literature. One algorithm is a linear programming-based algorithm that uses historical data to approximate customer arrival patterns and generate shadow prices for inventories. The price of inventory can be introduced to existing recommendation algorithms to obtain adjusted rankings for recommendation. The other algorithm balances expected revenue and inventory consumption, and it uses a simple penalty function to reduce the chance of recommending low-inventory-level products. Both algorithms are suitable for online recommendation systems for grocery stores with both online and offline channels, and can incorporate the features of perishable products, which need to be sold within limited time. Both algorithms are tested in a simulation using estimated parameters from Freshippo, a supermarket owned by the Alibaba Group. The numerical results show that both algorithms can generate higher sales volume and higher revenue.

We propose a novel method to extract Brillouin frequency shift (BFS) from Brillouin Gain Spectrum (BGS) in Brillouin distributed fiber sensors. The method is based on machine learning of nearest neighbors. In order to find the BFS from the BGS, we design two datasets, one for storing all possible BGS, and the other for storing the corresponding BFS. By comparing the given BGS with the dataset of BGS, we get the minimal kth BFS. The BFS of the given BGS is determined by voting of the kth BFS. By simulations, we compare the performance of both neighbor-based machine learning and curve-fitting. The results show that the method of neighbor-based machine learning is more robust under a wide range of signal-to-noise ratios, pump pulse widths, and frequency scanning steps. The extracting method of neighbor-based machine learning is highly competitive for future Brillouin distributed fiber sensors.

Advertisement, as a major revenue source of e-commerce platforms, is an important online marketing tool for sellers thereof. In this paper, we explore the dynamic ad allocation with limited slots upon each customer arrival for e-commerce platforms when the advertisers specify, for each ad, the budget constraint, the time periods when the ad should be displayed, and the click-through (lower-limit) constraint for certain customer segments. The goal of the platform is to maximize its payoff over the entire horizon. We propose a two-stage stochastic program framework in which the platform first decides the click-through goals for each ad/customer-type pair, and then devises the ad allocation policy to satisfy these goals in the second stage. We show that the optimal click-through goals can be achieved efficiently by solving a convex program, which can further reduce to a scalable linear program if the customer click-through behavior follows the multinomial logit model. Moreover, we provide a family of debt-weighted algorithms to achieve the optimal click-through goals, and prove that they are asymptotically optimal when the problem size scales to infinity. Compared to choice-based linear programming and its variant, our approach has better scalability and can deplete the ad budgets more smoothly throughout the horizon, which is very much desirable for the online advertising business in practice.

In this paper, combining with the tenth anniversary load test of Sutong Bridge, we proposed the Brillouin Optical Time Domain Analysis (BOTDA) to conduct health monitoring research on the bridge, and completed the following three aspects: Firstly as an auxiliary measure of strain monitoring, monitor the level of stress and strain along the longitudinal bridge to each position. Secondly through the parking of the vehicles at different positions on the bridge pavement, the influence lines on the overall structural stress of the bridge are obtained. Thirdly through the long-term strain monitoring of the bridge, the impact of the usual traffic load information on the bridge strain is obtained.The results show that the technology not only breaks through the monitoring bottleneck of traditional point sensors, but also realizes distributed measurement of strain on the transmission path; it can also be used for real-time monitoring, damage identification, crack location, settlement monitoring of traffic load information of bridges and other structures.. Compared with the traditional sensor test results, the effectiveness and frontier of the technology are proved, and the significance of the distributed fiber-optic sensor technology for the health monitoring of major structures such as bridges is fully explained.

An underwater thermal glider is efficient when it takes a stable zigzag-gliding in the vertical plane. To increase its endurance, improving the control and energy efficiency over the pitch angle adjustment is necessary. Firstly, the working and control principle of the thermal glider is given, and the mathematical motion model is built on this principle for a performance test. Secondly, based on the Tracking Differentiator and Active Disturbance Rejection Control (ADRC) theory, a self-searching optimal control method is proposed. With this method, the external disturbance can be compensated by adjusting the reference in advance, which makes the required control signal optimized and small. Finally, this method is used for the pitch-keeping control of the glider during the descent and ascent, or the posture transition control between the descent and ascent. The results show that the control system has a good steady and dynamic performance under a simulated ocean current and noise disturbance, as well as an adaptive effect. The control signal to keep and track the trajectory is much smaller. Thus, by assistantly adjusting the reference in advance, this method takes much less energy to resist the disturbance than by only changing the control output and adjusting the actuator with the ADRC or PID. Since the thermal glider carries a limited power, saving energy means an increased endurance. Theory is a guide before practice. This will have a great significance in the marine survey or military usage. Besides, the control algorithm is also simple. All these make it a better control strategy.

In recent years, empowered by rich media generation devices and convenient Internet access, Crowdsourced Live Game Video Streaming (CLGVS) has become one of the most popular Internet services. Twitch.tv, the most well-known CLGVS platform in the world, allows gamers to broadcast their gaming videos over the Internet. With the prevalence of mobile devices, viewers can watch gamers playing video games anywhere, anytime, on any devices (e.g., smartphones, tablets, or personal computers). However, the heterogeneity of user devices makes conventional solutions hard to ensure user-perceived quality. In this paper, we address the problem of cost-effective adaptive live game video streaming from the perspective of CLGVS service providers. Our purpose is to minimize the operational cost for CLGVS service providers by making live transcoding decisions, bit-rate adaptation decisions, and datacenter assignment decisions dynamically. Meanwhile, our algorithm also ensures good-enough service quality for viewers. Due to the diversity of game genres, we also consider game genres when designing our algorithm. To achieve the above purpose, we formulate the problem into a constrained stochastic optimization problem. By leveraging the Lyapunov optimization framework, we derive the online strategy with provable performance bound. To evaluate the effectiveness of our proposed algorithm, we further conduct a series of trace-driven simulations. The experimental results demonstrate the effectiveness of our algorithm in terms of operational cost and service quality. Our proposed algorithm can reduce operational cost by up to 50% while achieving good-enough viewer QoE compared with other alternatives.

In this paper, we study a multisourcing supply network design problem, in which each retailer faces uncertain demand and can source products from more than one distribution center (DC). The decisions to be simultaneously optimized include DC locations and inventory levels, which set of DCs serves each retailer, and the amount of shipments from DCs to retailers. We propose a nonlinear mixed integer programming model with a joint chance constraint describing a certain service level. Two approaches—set-wise approximation and linear decision rule-based approximation—are constructed to robustly approximate the service level chance constraint with incomplete demand information. Both approaches yield sparse multisourcing distribution networks that effectively match uncertain demand using on-hand inventory, and hence successfully reach a high service level. We show through extensive numerical experiments that our approaches outperform other commonly adopted approximations of the chance constraint.

Single-cell technologies are becoming increasingly widespread and have been revolutionizing our understanding of cell identity, state, diversity and function. However, current platforms can be slow to apply to large-scale studies and resource-limited clinical arenas due to a variety of reasons including cost, infrastructure, sample quality and requirements. Here we report DNBelab C4 (C4), a negative pressure orchestrated, portable and cost-effective device that enables high-throughput single-cell transcriptional profiling. C4 system can efficiently allow discrimination of species-specific cells at high resolution and dissect tissue heterogeneity in different organs, such as murine lung and cerebral cortex. Finally, we show that the C4 system is comparable to existing platforms but has huge benefits in cost and portability and, as such, it will be of great interest for the wider scientific community.

This paper presents a multi-objective optimization problem trade-off problem with a hybrid genetic algorithm to a construction project in China. The objective functions in this paper are to minimize the project time, to minimize the penalty cost, and to minimize carbon emissions. After describing the procedure of the problem in the project, this paper gives the mathematical formulation model. Furthermore, because traditional optimization techniques could not cope with the problem effectively, this paper presents a new approach based on the hybrid genetic algorithm (hGA). An adaptive hybrid genetic algorithm is developed for finding feasible solutions. Finally, a construction project is used as a practical example to demonstrate the practicality and efficiency of the model. Results are presented to highlight the performances of the optimization method, which is very effective and efficient as compared to other algorithms.

The electric energy alternative is conducive to improving energy efficiency and promoting the development of clean energy at the same time. The implementation of electric energy alternative can improve the level of electrification and the proportion of clean energy to terminal energy, which is the future development trend. This paper first analyzes the necessity of implementing electric energy alternative. Then the main influencing factors of electric energy alternative are analyzed. The potential evaluation model of electric energy alternative is also proposed. Finally, the potential evaluation model of electric energy alternative is analyzed deeply. The research results can play a good guiding role in the future of Chinese electric energy alternative work.

Under the guidance of the concept of green, low-carbon, energy-saving, environmental protection, and sustainable development, combined with the literature review at home and overseas and research on low-carbon construction in Chinese universities, this essay proposes a low-carbon campus evaluation index system. This system includes 5 first-level indicators and 17 second-level indicators from five dimensions, namely low-carbon management, low-carbon construction, low-carbon education, low-carbon research, and low-carbon life. In addition, this essay constructs a low-carbon campus evaluation model through using the fuzzy comprehensive analytic hierarchy process (AHP). Finally, it evaluates the construction of low-carbon campus in Chongqing Technology and Business University through using this model and verifies the validity of the model.

This paper presents a time-cost-carbon emissions trade-off problem with a hybrid genetic algorithm, to a large-scale construction project in China. The objective functions in this paper are to minimize the total project time, to minimize the total penalty cost, and to minimize carbon emissions. After describing the working procedure of the problem in the project, it gives the mathematical formulation model. Furthermore, because traditional optimization techniques could not cope with the problem effectively, we present a new approach based on the hybrid genetic algorithm (hGA). An adaptive hybrid genetic algorithm is developed for finding feasible solutions. Finally, a large scale construction projects is used as a practical example to demonstrate the efficiency and practicality of the model. Results are presented to highlight the performances of the optimization method, which is very efficient and effective as compared to other algorithms.

A206 at the conflict of speed and reliability in the transient stability assessment of the power system, this paper constructs a improved artificial neural network(ANN) classifier of transient stability, improving the speed of calculation by avoiding the unnecessary whole-process simulation. Combining the performance indexes gained by short-time simulation, the classifier predicts the transient stability and terminates the simulation in advance. Different from traditional classifier, this paper defines an overlap region to avoid misclassification. Finally, the application in the New England power system with 10 machines and 39 nodes proves the effectiveness and reliability of this ANN classifier.

This paper presents a design of the low noise induction magnetometer (IM) to measure the stray magnetic field for online monitoring of the induction machine. The frequency response and noise model of the IM are analyzed. An optimal IM is designed and developed, with the linear frequency response from 1Hz to 10kHz, which makes the IM be capable of time-domain and frequency-domain applications. The noise level of the IM is 2pT/VHz at 10Hz, 0.7fT/VHz at 20kHz. The magnetic fields during the safe and fault conditions of the induction generator are tested. The measurement results confirm that the low noise IM is efficient for online monitoring of the induction machine.

The penetration of mobile phones is nearly saturated in both developing and developed regions. In such a circumstance, how to prevent subscriber churn has become an important issue for today’s telecom operators, as the cost to acquire a new subscriber is much higher than that to retain an existing subscriber. In this paper, we propose to leverage the power of big data to mitigate the problem of subscriber churn and enhance the service quality of telecom operators. As the information hub, telecom operators have accumulated a huge volume of valuable data on subscriber behaviors, service usage, and network operations. To enable efficient big data processing, we first build a dedicated distributed cloud infrastructure that integrates both online and offline processing capabilities. Second, we develop a complete churn analysis model based on deep data mining techniques, and utilize inter-subscriber influence to improve prediction accuracy. Finally, we use real datasets obtained from a large telecom operator in China to verify the accuracy of our churn analysis models. The dataset contains the information of over 3.5 million subscribers, which generate over 600 million call detail records (CDRs) per month. The empirical results demonstrate that our proposed method can achieve around 90% accuracy for T + 1 testing periods and identify subscribers with high negative influence successfully.

We consider the capacity portfolio investment problem with flexible machines facing multiple products and demand uncertainty. For the problem of maximizing the service level, we approximate the objective with the largest inscribing sphere and provide a linear program formulation. We show, under certain conditions, that the optimal flexibility configuration consists only of dedicated machines and machines capable of producing only two types of products. Our work not only strengthens the sufficiency of limited flexibility in capacity investment but also suggests a novel way for approximating flexible system design problems in a computationally tractable form.

DC distribution network topology includes: chain structure, two-terminal structure and ring structure. This paper firstly introduces a device of Flexible Alternative Current Transmission System: static synchronous series compensator with fault current limiting. DC distribution network with SSSC-FCL can develop the power quality of distribution network and improve the stability. When the fault occurs, the SSSC-FCL can effectively limit the short circuit current to ensure the safety of system. And then this paper also introduces the main advantages of the DC distribution network and the topology, basic control methods of SSSC-FCL. Finally put forward a DC distribution network topology structure with SSSC-FCL. With the use of PSCAD/EMTDC to model and simulate the DC distribution network, the result of the simulation verified the feasibility and effectiveness of the DC network topology structure with SSSC-FCL.

With the development of marine science, aeronautics and astronautics, energy, chemical industry, biomedicine and management science, many complex systems face the problem of optimization and control. Approximate dynamic programming solves the curse of dimensionality problem of dynamic programming, and it is a new kind of approximate optimization solution that emerges in recent years. Based on the analysis of optimization system, this paper proposes a nonlinear multi-input multi-output, online learning, and data-driven approximate dynamic programming structure and its learning algorithm. The method is achieved from the following three aspects: 1) the critic function of multi-dimensional input critic module of the approximate dynamic programming is approximated with a data-driven k-nearest neighbor method; 2) the multi-output policy iteration of the approximate dynamic programming actor module is calculated with an exponential convergence performance; 3) The critic and actor modules are learned synchronously, and achieve the online optimal and control effect. The optimal control for the longitudinal motion of a thermal underwater glider is used to show the effect of the proposed method. This work can lay a foundation for the theory and application of a nonlinear data-driven multi-input multi-output approximate dynamic programming method. It's also the consensus needs in optimization control and artificial intelligence of many scientific and engineering fields, such as energy conservation, emission reduction, decision support and operational management etc.

In this paper, Four action-dependent heuristic dynamic programming control methods are presented for nonlinear multi-input-multi-output system with different characters based on the topology principle. These four methods are the action-network extension method, the sub-network method, the cascaded action-network method and the combined method. The derivation procedure and computing formulas of these methods are also derived. In it, the action-network extension method is mainly used for the conditions where the multi-output variables have the same orders of magnitude and a naturally coupled relationship. The sub-network method can nearly be applied in all cases and can solve the problem that the multi-output variables have different orders of magnitude. The cascaded action-network method is utilized when the multiple input variables have explicit cascaded relationships. The combined method can be used to control some highly regarded systems. Thus, these four methods can almost be used to satisfy all the design requirements of the nonlinear multi-input-multi-output control systems. The latter can refer to and select these methods as well as formulas for their control systems according to the research results to achieve a better control effect.

Compared with the AC distribution network, the DC distribution network has the advantages of lower line losses, better power quality, higher reliability and being suitable for all kinds distributed generation and DC load access, etc. This paper firstly studies the models of PV and energy storage connected to the DC distribution network. Then, a new type of DC distribution network topology with hybrid DC circuit breaker is proposed. A coordinated control strategy is also proposed, which can control energy storage to improve the stability of the system. Finally, using PSCAD/EMTDC to model and simulate. The simulation results show that the proposed topology of DC distribution network and control strategy can quickly remove the fault, shorten the fault power failure time and ensure the safe and stable operation of the system, which greatly improves the operation efficiency and safety level of the power grid.

Large-scale wind power access may cause a series of safety and stability problems. Wind power forecasting (WPF) is beneficial to dispatch in advance. In this paper, a new extreme gradient boosting (XGBoost) model with weather similarity analysis and feature engineering is proposed for short-term wind power forecasting. Based on the similarity among historical days&rsquo, weather, k-means clustering algorithm is used to divide the samples into several categories. Additionally, we also create some time features and drop unimportant features through feature engineering. For each category, we make predictions using XGBoost. The results of the proposed model are compared with the back propagation neural network (BPNN) and classification and regression tree (CART), random forests (RF), support vector regression (SVR), and a single XGBoost model. It is shown that the proposed model produces the highest forecasting accuracy among all these models.

Data deduplication, as a compression method, has been widely used in most backup systems to improve bandwidth and space efficiency. As data exploded to be backed up, two main challenges in data deduplication are the CPU-intensive chunking and hashing works and the I/O intensive disk-index access latency. However, CPU-intensive works have been vastly parallelized and speeded up by multi-core and many-core processors; the I/O latency is likely becoming the bottleneck in data deduplication. To alleviate the challenge of I/O latency in multi-core systems, multi-threaded deduplication (Multi-Dedup) architecture was proposed. The main idea of Multi-Dedup was using parallel deduplication threads to hide the I/O latency. A prefix based concurrent index was designed to maintain the internal consistency of the deduplication index with low synchronization overhead. On the other hand, a collisionless cache array was also designed to preserve locality and similarity within the parallel threads. In various real-world datasets experiments, Multi-Dedup achieves 3–5 times performance improvements incorporating with locality-based ChunkStash and local-similarity based SiLo methods. In addition, Multi-Dedup has dramatically decreased the synchronization overhead and achieves 1.5–2 times performance improvements comparing to traditional lock-based synchronization methods.

The project scheduling problems for large-scale construction systems has long been recognized as a critical optimization problem. In this study, a multi-objective programming model is developed for tackling a time-cost-quality-environment trade-off scheduling problem. For solving the multi-objective programming model, a two-stage algorithm is proposed as a combined solution method. In the first stage, the weight-sum procedure is adopted to deal with the multi-objective model, therefore to ensure the validity of conformity in multi-objectives, dividing out the dimensions and unifying the order of magnitude must be performed before the weight-sum procedure; in the second stage, EBS-based GA is developed to solve the problem. Finally, the results and comparisons analysis of a case study are presented to demonstrate the practicality and efficiency of the optimization method.

This paper presents a discrete time-cost-environment trade-off problem (DTCETP) by a hybrid genetic algorithm, to a large-scale construction project in the southwest region of China. The objective functions in this paper are to minimize the total project time, to minimize the total penalty cost, and to minimize environmental impact. After describing the problem of the working procedure in the project, it gives the mathematical formulation model of a DTCETP. Furthermore, because traditional optimization techniques could not cope with the DTCETP effectively, we present a new approach based on the hybrid genetic algorithm (hGA). A hybrid genetic algorithm is developed for finding feasible solutions. Finally, a large scale construction projects is used as a practical example to demonstrate the practicality and efficiency of the model. Results and sensitivity analysis are presented to highlight the performances of the optimization method, which is very effective and efficient as compared to other algorithms. The computer generated results validate the effectiveness of the proposed model and algorithm in solving large-scale practical problems.

This paper proposed a new hybrid spectral clustering algorithm in which Mean Impact Value (MIV) was used in the cost dimension reduction. The processing of system implementation is as follows: first, we used BP neural network to determine the principal items materials, and then applied the spectral clustering algorithm to calculate the principal items price according to principal items materials; finally, principal items ratio estimation method has been used to do cost estimation. This paper took the Xiamen project cost station as the actual case and experimental results shown that this algorithm could meet the cost requirements of project cost station both in time efficiency and accuracy through parameters self-adjusting.

The concept of chaining, or in more general terms, sparse process structure, has been extremely influential in the process flexibility area, with many large automakers already making this the cornerstone of their business strategies to remain competitive in the industry. The effectiveness of the process strategy, using chains or other sparse structures, has been validated in numerous empirical studies. However, to the best of our knowledge, there have been relatively few concrete analytical results on the performance of such strategies vis-á-vis the full flexibility system, especially when the system size is large or when the demand and supply are asymmetrical. This paper is an attempt to bridge this gap. We study the problem from two angles: (1) For the symmetrical system where the (mean) demand and plant capacity are balanced and identical, we utilize the concept of a generalized random walk to evaluate the asymptotic performance of the chaining structure in this environment. We show that a simple chaining structure performs surprisingly well for a variety of realistic demand distributions, even when the system size is large. (2) For the more general problem, we identify a class of conditions under which only a sparse flexible structure is needed so that the expected performance is already within ϵ optimality of the full flexibility system. Our approach provides a theoretical justification for the widely held maxim: In many practical situations, adding a small number of links to the process flexibility structure can significantly enhance the ability of the system to match (fixed) production capacity with (random) demand.

One of the most effective ways of minimizing supply/demand mismatch costs, with little increase in operational costs, is to deploy valuable resources in a flexible and timely manner to meet the realized demand. This notion of flexible processes has significantly changed operations in many manufacturing and service companies. For example, a flexible production system is now commonly used by automobile manufacturers, and a workforce cross-training system is now a common practice in many service industries. However, there is a trade-off between the level of flexibility available in the system and the associated complexity and operating costs. The challenge is to have the “right” level of flexibility to capture the bulk of the benefits from a fully flexible system, while controlling the increase in implementation costs. This paper reviews developments in process flexibility over the past decade. In particular, we focus on the phenomenon, often observed in practice, that a slight increase in process flexibility can lead to a significant improvement in system performance. This review explores the issues from three perspectives: design, evaluation, and applications. We also discuss how the concept of process flexibility has been deployed in several manufacturing and service systems.

Production postponement, the strategy to hold reserved production capacity that can be deployed based on actual demand signals, is often used to mitigate supply-demand mismatch risk. The effectiveness of this strategy depends crucially on the ease, or flexibility, in deploying the reserved capacity to meet product demands. Existing literature assumes that the reserved capacity is fully flexible, i.e., capable of being deployed to meet the demand of any item in a multiproduct system. Little is known if reserved capacity is held at many different locations, with each location having only a limited range of flexibility on production options. This paper examines how effective the production postponement strategy is in this environment. When the amount of reserved capacity is small (i.e., postponement level near 0%), no amount of flexibility can reap significant benefits. When the reserved capacity is high (i.e., postponement level near 100%), it is well known that a sparse structure such as a 2-chain can perform nearly as well as a fully flexible structure. Hence, process flexibility beyond 2-chain has little impact on the effectiveness of production postponement strategy in these two extreme environments. Interestingly, in a symmetric system, we prove that the performance of 2-chain, vis-à-vis the full flexibility structure, has a wider gap when postponement level (i.e., amount of reserved capacity) is moderate, and thus process flexibility beyond 2-chain matters and affects appreciably the performance of the production postponement strategy. Fortunately, adding a little more flexibility, say turning a 2-chain into a 3-chain, the system can perform almost as well as a full flexibility structure for all postponement levels. This is important as first stage production capacity can be allocated as if the reserve capacity is fully flexible. Our analysis hinges on an exact analytical expression for the performance of d-chain, obtained from solving a related class of random walk problems. To the best of our knowledge, this is the first paper with analytical results on the performance of d-chain for d > 2. ASTAR (Agency for Sci., Tech. and Research, S’pore) Accepted version

A multispectral camera acquires spectral color images with high fidelity by splitting the light spectrum into more than three bands. Because of the shift of focal length with wavelength, the focus of each channel should be mechanically adjusted in order to obtain sharp images. Because progressive adjustment is quite time consuming, the clear focus must be determined by using a limited number of images. This paper exploits the symmetry of focus measure distribution and proposes a simple yet efficient autofocus method. The focus measures are computed using first-order image derivatives, and the focus curve is obtained by spline interpolation. The optimal focus position, which maximizes the symmetry of the focus measure distribution, is then computed according to distance metrics. The effectiveness of the proposed method is validated in the multispectral camera system, and it is also applicable to relevant imaging systems.

With the rapid development of multispectral imaging tech- nique, it is desired that the spectral color can be accurately repro- duced using desktop color printers. However, due to the specific spectral gamuts determined by printer inks, it is almost impossible to exactly replicate the reflectance spectra in other media. In addition, as ink densities can not be individually controlled, desktop printers can only be regarded as red-green-blue devices, making physical models unfeasible. We propose a locally adaptive method, which con- sists of both forward and inverse models, for desktop printer charac- terization. In the forward model, we establish the adaptive transform between control values and reflectance spectrum on individual cellular subsets by using weighted polynomial regression. In the inverse model, we first determine the candidate space of the control values based on global inverse regression and then compute the optimal control values by minimizing the color difference between the actual spectrum and the predicted spectrum via forward transform. Experimental results show that the proposed method can reproduce colors accurately for different media under multiple illumi- nants. © 2013 SPIE and IS&T (DOI: 10.1117/1.JEI.22.2.023012)