Judgmental forecasting with interactive forecasting support systems

With an increasing use of DSS/EIS, managers are often required to process information coming from a variety of sources in making a final decision. However, we have little understanding of the efficiency with which people select and use the multiple pieces of information. This issue was examined under various conditions using a DSS in a forecasting task where multiple items of information were displayed on request in an interactive manner. Results indicate that overall people underacquired information. Moreover, people often selected less-reliable information. This sub-optimal behaviour did not diminish over time (it became worse). But an aggregation DSS was helpful at the task. This suggests that people seemed to have a problem in aggregating multiple pieces of information. It was also found that the independent preparation of an initial forecast improved forecast accuracy significantly. Perhaps, forecasters may prepare the initial forecast independently and use decision aids for the subsequent tasks of the forecasting process.     

Long‐Range Forecasting Highway Construction Costs

In the process of decision making for design and execution of highway construction projects, long‐range cost forecasting is one of the most significant and complicated problems. This paper describes the development of a model that enables the user to make long‐range cost projections, taking into consideration general characteristics of the highway construction industry, as well as pertinent local conditions. The model presented uses conventional statistical methods to represent the main categories of typical jobs in the highway construction industry. From these categories, a composite model is created by assigning different weights to the input elements costs and then choosing a series of indicators to predict price trends for each separate element of the composite model. Use of this model reveals that bid volume in a certain area is a factor that has significant influence upon cost forecasts. This paper is accompanied by a case study based on actual data from highway construction projects performed for the Florida Department of Transportation in the years 1968–1981.     

A non-parametric approach to demand forecasting in revenue management

In revenue management, the profitability of the inventory and pricing decisions rests on the accuracy of demand forecasts. However, whenever a product is no longer available, true demand may differ from registered bookings, thus inducing a negative bias in the estimation figures, as well as an artificial increase in demand for substitute products. In order to address these issues, we propose an original Mixed Integer Nonlinear Program (MINLP) to estimate product utilities as well as capturing seasonal effects. This behavioral model solely rests on daily registered bookings and product availabilities. Its outputs are the product utilities and daily potential demands, together with the expected demand of each product within any given time interval. Those are obtained via a tailored algorithm that outperforms two well-known generic software for global optimization.     

基于ARMA模型的双路径差异性分析

拟通过单向时延的历史记录对其未来趋势进行预测。首先,通过粗粒度和细粒度两种方式采集世界各地若干目标节点的单向时延作为原始数据;然后,结合多种预测模型的特点和原始数据的内在属性,提出一种基于ARMA模型的双路径差异性分析方法;最后选取合理的预测窗口对方法的准确性进行验证。结果表明,该预测方法是正确、合理的,与灰色预测模型相比能更有效地预测单项时延差序列。     

Kalman filter based production control of a failure-prone single-machine single-product manufacturing system with imprecise demand and inventory information

An adaptive production control structure for failure-prone manufacturing systems under inventory and demand uncertainty is proposed. It contains estimation and forecasting modules incorporated into a control loop. The customer demand is unknown and its rate is composed of ramp-type, seasonal and random components. Information available to decision maker consists of imprecise inventory records, and the Kalman filter technique is used for estimating the inventory level and demand rate online from noisy inventory measurements. Estimates obtained are shown to converge to the actual values in stochastic sense. They are subsequently used for demand component forecasting, once the estimation errors become sufficiently small. A forecasting algorithm allows estimating ramp-type and seasonal demand components, together with their potential errors. Obtained estimates are incorporated into production control procedures, recently developed for manufacturing systems under variable and uncertain demand. Optimality conditions in the form of Hamilton-Jacobi-Bellman equations are obtained. A constructive numerical method for computing sub-optimal production policies is proposed and validated through numerical simulations.     

Machine Learning Supports Robust Operation of Thermosiphon Reboilers

The analysis of process and equipment operational data in chemical engineering regularly requires a high level of expert knowledge. This work presents a Machine Learning-based approach to evaluate and interpret process data to support robust operation of a thermosiphon reboiler. By applying an outlier detection, potentially interesting and unstable operating conditions can be identified quickly. A multidimensional regression allows to forecast the circulating mass flow. The results obtained fit well into the current state of research and manual evaluation of thermosiphon reboilers.     

Competitive substitution and technological diffusion for semiconductor foundry firms

To survive in an intensively competitive environment, semiconductor companies need to be more agile, responsive, and flexible than before. Generally, semiconductor industry consists of three business models: integrated design manufacturer, fabless chip design, and foundry business. In general, semiconductor firms are affected by three drivers: new entrants and rivals (competition), main customers (demand), and process technologies (R&D). Inspired by Michael Porter’s five-force analysis, a novel framework is presented to accomplish the following goals: (1) The interactions between the top three foundry firms, TSMC, Samsung, and Global Foundries, are analyzed to reveal managerial insights, (2) The main customers of TSMC including Apple, Huawei, Qualcomm, Mediatek, AMD, and NVidia are incorporated to into sales forecasting, and (3) Technological diffusion across the mature process (above 70 nm), the medium process (between 20 nm and 70 nm), and the advanced process (below 20 nm) is captured to predict sales revenues. Key findings are shown as follows: (1) a large-scale foundry frim benefits from the existence of a small-scale firm, (2) the inclusion of main customers significantly improves the performance of sales forecasting, (3) the advanced process gradually benefits from the mature process while it rapidly replaces the medium process in a “predator-prey” way.     

基于数据挖掘技术的天然气价格预测方法研究

天然气价格是影响天然气企业经营决策与运营效益的重要因素,在此背景下,如何准确地预测未来天然气价格自然成为产业界关注的热点话题。此外,在数据挖掘技术快速发展的时代,如何将该技术应用于传统的天然气行业,融入天然气价格的预测当中,也是学术界所探讨的重要话题。基于此,本文首先回顾了以往天然气价格预测方法,然后以传统数据挖掘技术中的模式序列相似性搜索方法(PSS)为基础,通过对该方法中历史序列搜索匹配机制及结果处理机制的改进,提出了一种新的改进模式序列相似性搜索(APSS)天然气价格预测方法。在方法构建之后,采用美国天然气日度现货价格数据对该方法的有效性进行了实验验证。实验结果表明,本文提出的基于数据挖掘技术的APSS方法能够实现对天然气价格的合理预测,且与传统的PSS方法相比,APSS方法的预测结果具有更高的预测精度。     

外圆无心磨削质量控制的灰色预测模型

本文对如何建立外圆无心磨削质量控制的灰色预测模型问题进行了探讨，重点分析了等维灰数递推动态预测模型的建模方式，也初步分析了灰区间预测控制和灰模块预测控制的可能性，通过实验验证了预测的精度．     

一种基于人工神经网络的短期负荷预测技术

提出了一种基于人工神经网络的电力系统短期负荷预测技术,该方法在计及气温因素对负荷预测影响的基础上,将神经网络同一种较为新颖的预报模型相结合,因而具有较高的预测精度。计算实例证明了该法的可行性和有效性。