Optimisation and heuristic approaches for assigning inbound containers to outbound carriers

Due to economical and/or geographical constraints, most of the time overseas containers cannot be directly shipped to their destinations. These containers visit transhipment ports where they are first unloaded and temporarily stored and then loaded onto smaller vessels (feeders) to be transported to their final destinations. The assignment of these containers to outbound vessels necessitates several factors to be taken into account simultaneously. In this paper, we develop a mathematical model to reflect multiple objectives with priorities and to assign these containers to different vessels at the transit container port terminal. Although we solve a single-objective (with the weighted sum of objectives) mathematical model to optimality, we also propose two heuristic approaches to solve this complex problem for a transit agency. The first heuristic is shipment based and has four variants differing in how the opportunity costs of the assignments are calculated. The second greedy heuristic is trip based, where the goal is to maximise the capacity utilisation of the vessels. The heuristics return very promising solutions in ignorable computational times. We also provide real-life cases and present our conclusions.     

Timetable optimization models and methods for minimizing passenger waiting time at public transit terminals

This paper focuses on developing mathematical optimization models for the train timetabling problem with respect to dynamic travel demand and capacity constraints. The train scheduling models presented in this paper aim to minimize passenger waiting times at public transit terminals. Linear and non-linear formulations of the problem are presented. The non-linear formulation is then improved through introducing service frequency variables. Heuristic rules are suggested and embedded in the improved non-linear formulation to reduce the computational time effort needed to find the upper bound. The effectiveness of the proposed train timetabling models is illustrated through the application to an underground urban rail line in the city of Tehran. The results demonstrate the effectiveness of the proposed demand-oriented train timetabling models, in terms of decreasing passenger waiting times. Compared to the baseline and regular timetables, total waiting time is reduced by 6.36% and 10.55% respectively, through the proposed mathematical optimization models.     

The multi-objective railway timetable rescheduling problem

Unexpected disruptions occur for many reasons in railway networks and cause delays, cancelations, and, eventually, passenger inconvenience. This research focuses on the railway timetable rescheduling problem from a macroscopic point of view in case of large disruptions. The originality of our approach is to integrate three objectives to generate a disposition timetable: the passenger satisfaction, the operational costs and the deviation from the undisrupted timetable. We formulate the problem as an Integer Linear Program that optimizes the first objective and includes $\epsilon$-constraints for the two other ones. By solving the problem for different values of $\epsilon$, the three-dimensional Pareto frontier can be explored to understand the trade-offs among the three objectives. The model includes measures such as canceling, delaying or rerouting the trains of the undisrupted timetable, as well as scheduling emergency trains. Furthermore, passenger flows are adapted dynamically to the new timetable. Computational experiments are performed on a realistic case study based on a heavily used part of the Dutch railway network. The model is able to find optimal solutions in reasonable computational times. The results provide evidence that adopting a demand-oriented approach for the management of disruptions not only is possible, but may lead to significant improvement in passenger satisfaction, associated with a low operational cost of the disposition timetable.     

HTS磁体的失超判别

高温超导磁体的失超过程比较缓慢,失超判别很困难。本文提出定性的失超判别方法：在电流恒定的前提下,若磁体电压持续上升,则认为磁体失超。对试验数据进行线性拟合,提出失超判别的数学表示形式：磁体电压的变化率dV／dt〉0。     

一种考虑港城互动发展的线性回归模型

如果自变量确定,直接建立自变量x和因变量y的关系y=f（x）,则可以很好地预测y。而在自变量与因变量都不确定的情况下,建立x与y的关系y=f（x）,或者x=g（y）都不合理。本研究提出一种新的预测方法,即建立综合推导公式来预测x与y。通过实例证明,此公式能使x与y相互推导的总误差最小,是可行的。     

机场停机位指派模型及算法

为探索中小型机场停机位指派问题有效合理的解决方法,以乘客最小行走距离为优化目标,建立了中小型机场停机位整数规划指派模型,利用改进的启发式算法进行了求解。该算法可以缩小搜索空间,在最坏情况下的计算复杂度是多项式的,优于CPLEX软件的指数级计算复杂度,具有更高的计算效率;与CPLEX结果相比,距离误差仅为5.2%,指派结果令人满意。     

基于位移与轴力的软土深基坑开挖扰动控制

针对软土地层基坑工程钢支撑体系压力控制液压伺服反馈控制中施工扰动最终位移的不确定性及其环境风险,研究建立了以扰动位移控制为核心、位移与支撑轴力综合优化的钢支撑液压伺服智能控制方法。提出基于容许位移约束的线性位移变化与支撑力综合优化智能控制模式,并建立相应的施工全过程实时调压技术。构建了基于高精度激光测距和液压监测同步的反馈系统,采用形函数的随机数据误差处理方法,有效防范随机误差带来的反馈控制失误和风险。经模型试验,验证了该方法的可靠性和适用性。经上海世博通道工程示范应用,位移控制效果显著。     

柱式桥墩盖梁内力计算方法探讨

笔者阐述了两种不同计算图式下盖梁内力计算方法,并进行了对比分析.分析表明,当考虑立柱刚度影响时,盖梁内力将进行二次分配,致使跨中正弯矩减小、支点负弯矩增大(称之为“刚架效应”),随着立柱与盖梁的线刚度比的增大,此种效应越显著.当立柱与盖梁的线刚度比较小时,可采用简化的双悬臂简支梁(连续梁)图式,但在支点截面应作适当的加强处理;而对于重要桥梁或立柱与盖梁的线刚度比较大的桥梁,宜采用双悬臂刚架结构图式进行计算.     

埋地管道外覆盖层绝缘电阻率计算及应用

外覆盖层是埋地钢质管道免遭外界腐蚀的第一道防线,其保护效果直接影响着管道整体的运行效率,而管道外覆盖层电阻率是评价外覆盖层状况好坏的最主要因素.运用线传输理论,通过将管地系统简化为一个集中参数等效电路,建立了埋地钢质管道外覆盖层绝缘电阻率的计算模型,并给出了管地系统参数的具体确定方法,开发了外覆盖层绝缘电阻率的计算软件.利用PCM检测数据,基于建立的Rg计算方法,能够实现对Rg值的计算.实践表明:计算结果能够很好地反映实际情况,具有一定的工程应用价值.