Alliance and non-alliance modes are the typical decision-making problem of strategy layer. Upper layer decision will have direct impact on the lower two layers: tactical layer and operational layer. Empty container repositioning mechanism must be keep synchronous operation according to the upper two layers. To quantitative analyse the impact of two modes of strategy layer, we have built the optimization models of empty containers repositioning according to the actual business flow. Two simulation cases have been designed to solve the models. The experimental data show that under alliance it can reduce various costs dramatically and improve the profits contrast to the non-alliance. The results validate the proposed standpoint: alliance mode is feasible for different company sizes. All companies in consolidation can obtain win–win. Furthermore, the reduced various costs such as transportation cost also mean lower greenhouse gas emissions, which is importance to keep sustainable development for the earth.

Due to the special role of empty containers in the container transportation process, empty container repositioning is a focal point in the shipping industry. For this problem, highly efficient and feasible optimization models are critical in improving the benefit for shipping companies and increasing their market competitiveness. Operational decisions are affected by tactical ones. Aimed at this point, we propose a tactical and operational cooperative empty container repositioning optimization model. To cut the search space and obtain the optimal solution quickly, several initial solutions generation rules are extracted, based on business flow. Furthermore, the reachable shipping distance may change when the calling sequence is different. An algorithm which calculates the reachable shipping distance matrix between ports is presented to solve this problem. Simulated cases are used to test the proposed model and algorithm. The results show that the cases can cope with the tactical and operational cooperative empty container repositioning optimization model. Moreover, some interesting conclusions also are deduced about the relationships among number of calling ports, total profits, leasing cost, calling port fee, number of Empty Containers Repositioned (ECR), and laden containers. All these can guide and assist the various decisions to be made. According to the homepage of Symmetry, its subject areas include Mathematics, Computer Science, Theory, and Methods. Their branches include information theory, computer-aided design, and so on. The topic of our paper is to solve this engineering application problem by using a mathematical optimization model and computer methods. That is, applying mathematical theory and computer methods to make decision results for the empty container repositioning problem in the shipping industry. It has certain economic value and practical significance. Obviously, it is consistent with the theme of Symmetry.

Compared with ordinary image classification tasks, fine-grained image classification is closer to real-life scenes. Its key point is how to find the local areas with sufficient discrimination and perform effective feature learning. Based on a bilinear convolutional neural network (B-CNN), this paper designs a local importance representation convolutional neural network (LIR-CNN) model, which can be divided into three parts. Firstly, the super-pixel segmentation convolution method is used for the input layer of the model. It allows the model to receive images of different sizes and fully considers the complex geometric deformation of the images. Then, we replaced the standard convolution of B-CNN with the proposed local importance representation convolution. It can score each local area of the image using learning to distinguish their importance. Finally, channelwise convolution is proposed and it plays an important role in balancing lightweight network and classification accuracy. Experimental results on the benchmark datasets (e.g., CUB-200-2011, FGVC-Aircraft, and Stanford Cars) showed that the LIR-CNN model had good performance in fine-grained image classification tasks.

The problem of empty containers repositioning (ECR) is to dispatch the empty containers coupled with laden containers transportation flow between the surplus ports and the deficit ports according to the fixed schedule. Because this problem involves many parameters, variables and constraints, how to build and solve the model is always focus point. Aiming at this problem, firstly one loop itinerary of liner is divided into several stages to analyze the changes of two typical containers located in the ports and on board with the time evolving. Secondly, for attacking the difference between supply and demand empty containers in preview stage t and the ones in next stage t + 1, one dynamic region-to-region ECR model, which adopts the dynamic across-region port set to port set redistribution strategy, is proposed to reduce the various possible costs which commonly exist in existing static port-to-port ECR policy. Finally, through some deductive instances and contrast model we analyze these two models. Results show that in any case our proposed model has absolutely more advantages than the one who uses the static port-to-port ECR strategy. Moreover, the former has significant information support for making the safety stock of ports.