Volume 8, Issue 3

Linear and mixed integer programming are very popular and important methods to make efficient scientific management decision. With large size of real application data, the use of linear-mixed integer programming is facing problems with more complexity; therefore, preprocessing techniques become very important. Preprocessing aims to check and delete redundant information from the problem formulation. It is a collection of techniques that reduce the size of the problem and try to strengthen the formulation. Fast and effective preprocessing techniques are very important and essential for solving linear or mixed integer programming instances. In this paper, we demonstrate a set of techniques to presolve linear and mixed integer programming problems. Experiment results showed that when preprocessing is well done, then it becomes easier for the solver; we implemented interior-point algorithm for computational experiment. However, preprocessing is not enough to reduce the size and total nonzero elements from the constraints matrix. Moreover, we also demonstrate the impact of minimum degree reordering on the speed and storage requirements of a matrix operation. All techniques mentioned above are presented in a multifunctional software to facilitate users.

With the boom of web technology, Internet concerns (IC) have become emerging drivers of crude oil price. This paper makes the first attempt to measure the frequency-varying co-movements between crude oil price and IC in five domains (i.e., fundamentals, supply-demand, crisis, war and weather) by using the frequency causality test method. Based on the monthly Brent spot price and search volumes (SVs) captured by Google Trends from January 2004 to September 2019, new and complementary insights regarding the co-movements between crude oil price and IC are obtained. 1) The co-movements between crude oil price and the IC of supply-demand, war, and weather support a neutral hypothesis at all frequencies due to the characteristics (low value or volatility) of these IC data. 2) There is a unidirectional causal relationship between crude oil price and the IC of fundamentals, running from the latter to the former at low frequencies (long-term). 3) There is a feedback relationship between crude oil price and the IC of crisis, with the IC of crisis driving crude oil price at medium and low frequencies (mid- and long-term) and crude oil price causing the IC of crisis to change permanently. The conclusions of this paper provide important implications for both oil market economists and investors.

In this paper, M-Copula is used to analyze the correlation between Shanghai Composite Index and Shanghai Fund Index. By analyzing the characteristics of the logarithmic yields sequence of two samples, the marginal distribution model is established by using EGARCH-M-GED model. According to the correlation between two logarithmic yields sequence, the M-Copula model is selected to model its correlation structure, and its parameters are estimated by EM algorithm. Because M-Copula combines characteristics of different Copulas, it has more flexible distribution forms and more prominent ability to describe the fat tails and correlation characteristics of data, and more importantly, the effect is better than single Copula.

Truck and trailer routing problem (TTRP) is one of the most frequently encountered problem in city distribution, particularly in populated and intensive downtown. This paper addresses this problem and designs a novel backtracking search algorithm (BSA) based meta-heuristics to solve it. The initial population is created by T-sweep heuristic and then based on the framework of backtracking search algorithm, four types of route improvement strategies are used as building blocks to improve the solutions of BSA in the process of mutation and crossover. The computational experiments and results indicate that the proposed BSA algorithm can provide an effective approach to generate high-quality solutions within the satisfactory computational time.

An M / M /1 constant retrial queue with balking customers and set-up time is considered. Once the system becomes empty, the server will be turned down to reduce operating costs, and it will be activated only when there is a customers arrives. In this paper, the almost unobservable case is studied, in which the information of the queue length is unavailable, whereas the state of the server can be obtained. Firstly, the steady state solutions are derived and the individual equilibrium strategies are analyzed. In addition, social optimization problems, including cost analysis and social welfare maximization are investigated by using the PSO algorithm. Finally, by appropriate numerical examples, the sensitivity of some main system parameters is shown.