Search

It is imperative for manufacturing systems to improve production quality not only to maintain profitability, market share and competitiveness, but also to reduce energy waste resulted from defective items. Although quality and energy saving have attracted extensive attention in the past few decades, there is little research effort devoted to a systematic understanding of their intersection. Therefore, this paper analyzes the energy usage of a multistage production system with quality control. The Geometric reliability and Bernoulli quality models are assumed. A Markov process model is established to predict the dynamics of the production system. The energy economics of the production system are analyzed to include both production and energy cost. The optimal PWQ (Production with Quality Inspection) machine allocation method and the cost-effectiveness analysis method are formulated to increase the profit. According to the computational experiments, the proposed optimal PWQ machine allocation method can effectively reduce energy consumption. In addition, pay-back period (PBP) is an effective indicator which helps production managers make cost effective decisions for machine replacement. The research results in an in-depth understanding of the energy economics of systems with quality control, which is necessary for manufacturers to gain competitiveness with better product quality and higher energy efficiency.

Cloud manufacturing (CMfg) can make mass manufacturing resources more widely accessible to users in the form of manufacturing services with their available time slots and the associated prices. During the traditional service process, clients do not have any mutual communication or interaction. This paper proposes a new operational mode for CMfg in which clients can coordinate with each other for the exchange and reutilization of their booked resources. In the client coordination mode, a centralized optimization is conducted for clients to develop an overall coordinated solution, which creates potential cost savings compared to traditional uncoordinated solutions. The optimization problem, which is a joint service booking and scheduling problem, has been proved to be strong NP-hard. Besides, we consider manufacturing synchronization (MfgSync) when booking services for clients, which includes simultaneity of completing the sub-tasks of a client's manufacturing job and punctuality of fulfilling the job. We propose two optimal solution algorithms based on dynamic programming to address the optimization problem for two common cases. Considering the selfishness of each client, the success of the coordination depends on whether every participant can be better off than to act independently. We develop a cooperative game for the client coordination mode and propose a core cost savings allocation scheme with which every client can get benefits through coordination. Based on the allocation scheme, clients tend to form a grand coalition. The computational results show that our approach can make substantial cost savings for clients.

The batching machine can process several jobs simultaneously as a batch. Aiming at the Bernoulli serial production lines with finite buffers and batching machines, we establish an analytical model for the two-machine one-buffer Bernoulli serial lines and an approximate method for the longer lines to evaluate the performance measures. Our numerical experiments verify the high accuracy of such methods.

Shifting economic bottleneck identification aims to improve the overall throughput of manufacturing system. Firstly the two new concepts, throughput sensitivity of machine and corresponding economic bottleneck, are mathematically defined. Secondly considering both throughput effect and active duration of machine, a new two-dimensional bottleneck indicator named active throughput and duration (ATD) is presented and new concave-convex Gantt chart is given for scheduling representation. Thirdly the shifting economic bottleneck identification (SEBI) method is presented to identify either the momentary economic bottleneck or statistically stable bottleneck. What's more, it also works well when facing the sole economic bottleneck, multiple economic bottlenecks and shifting economic bottleneck. Finally a practical job shop scheduling problem is optimized and the bottleneck identification is gained by the proposed method. Also a comparison between the proposed method and existing method is made. The results show that the proposed approach is effective and explicit for shifting economic bottleneck identification.

We address the problem of pedestrian detection in still images. Current pedestrian detection systems are hard to improve both speed and accuracy simultaneously. In order to achieve a balance between speed and accuracy, we propose a novel MPL-Boosted cascade of weak FIK-SVM classifiers. Our method achieves high recall while taking the speed-advantage of cascade-of-rejectors approach. Each feature in our algorithm corresponds to a 66-D HOG-LBP feature vector that describe a block. The weak classifiers we use are the separating hyper-plane computed by using a FIK-SVM. We use MPL-Boost to select features from a large set of possible blocks. The integral image and convoluted trilinear interpolation are used for rapid calculation of block feature. For a 320×240 image, the system can process 16 frames per second with sparse scan, while defeat the accuracy level of existing methods.

Numerous digital cameras and modern phones have a face detection module, which is used to automatically focus (AF) and optimize exposure (AE). But the face detection will fail when person doesn't face the camera or the part of the face is occluded. In order to avoid such problems, we propose a fast head-shoulder detector, which uses Variable-size block Histograms of Orientated Gradients (VHOG) descriptors. AdaBoost-based feature selection algorithm and integral image representation are used to speed up the algorithm. The tests reveal that the method shows very good results and works efficiently in spite of the low computational power and memory available in mobile devices.

A new heuristic algorithm for large-scale job shop scheduling problem is proposed based on operation decomposition using bottleneck machine. It detects bottleneck machine by orthogonal experiment. Based on the operations which are processed on the bottleneck machine, the large-scale job shop scheduling problem is divided into three sub-problems: the scheduling problem of the operations on bottleneck machine, pre-bottleneck machines and post-bottleneck machines. The operations on bottleneck machine are first scheduled by dispatching rules, and the other two sub-problems are solved by genetic algorithm to subordinate the schedule of the operations on bottleneck machine. A number of simulation results shown that the scheduling strategy and technique for the problem can improve the computing efficiency and quality of the solution.

Aiming at China's aviation industrial enterprises, a reconfigurable manufacturing execution system (RMES) is presented to meet their changeable and individual requirements. Firstly, based on the analysis of aviation product characteristics and production management mode, the existing problems are analyzed. Secondly, a MES solution named Workshop Manager for China's aviation industrial enterprises is presented. Here, Workshop Manager is divided into 15 subsystems, such as planning, scheduling, inventory, tool, quality, equipment, cost management, with their functions being specifically designed, and their business processes being optimized as well. Moreover, a four-tier MES architecture based on SOA is established. Afterwards, an information integration of workshop manager with other hardware and software systems is discussed in details. Finally, the Workshop Manager is developed successfully by employing component technology. The application results within more than 30 job-shops in 7 large and middle-sized enterprises of China have confirmed that the integrated and reconfigurable Workshop Manager is feasible and will be rather effective.

A new bottleneck definition from the perspective of the manufacturing system’s objective is proposed. A distinct bottleneck detection method based on orthogonal experiment (BD-OE) for job shop is proposed according to the definition. The method takes the objective of the manufacturing system as estimated index, carries out typical and finite number of experiments by combining different dispatching rules, to detect bottleneck machine which has the greatest effect on the estimated index. The method can detect bottleneck before the system running, and to guide the following production planning and preparation of necessity in advance for the improvement of the system’s performance.

International audience; The job-shop scheduling problem with discretely controllable processing times (JSP-DCPT) is formulated. A decomposition strategy is applied so that the problem could be solved by solving a classical JSP problem and a series of discrete time-cost tradeoff problems. Based on the partition of the solution space and the appliance of the set dominance, an extreme mode based set dominant decomposition (EMSDD) algorithm is proposed. It is proved that the solution space of JSP-DCPT could be reduced by restricting the makespan to an upper bound value which is derived by EMSDD.