Search results

Intermittent demand appears sporadically, with some time periods not even displaying any demand at all. Even so, such patterns constitute considerable proportions of the total stock in many industrial settings. Forecasting intermittent demand is a rather difficult task but of critical importance for corresponding cost savings. The current study aims to examine the empirical outcomes of three heuristics towards the modification of established intermittent demand forecasting approaches.

First, optimization of the smoothing parameter used in Croston's approach is empirically explored, in contrast to the use of an a priori fixed value as in earlier studies. Furthermore, the effect of integer rounding of the resulting forecasts is considered. Lastly, the authors evaluate the performance of Theta model as an alternative of SES estimator for extrapolating demand sizes and/or intervals. The proposed heuristics are implemented into the forecasting support system.

The experiment is performed on 3,000 real intermittent demand series from the automotive industry, while evaluation is made both in terms of bias and accuracy. Results indicate increased forecasting performance.

The current research explores some very simple heuristics which have a positive impact on the accuracy of intermittent demand forecasting approaches. While some of these issues have been partially explored in the past, the current research focuses on a complete in‐depth analysis of easy‐to‐employ modifications to well‐established intermittent demand approaches. By this, the authors enable the application of such heuristics in an industrial environment, which may lead to significant inventory and production cost reductions and other benefits.

Solar power forecasting will have a significant impact on the future of large-scale renewable energy plants. Predicting photovoltaic power generation depends heavily on climate conditions, which fluctuate over time. In this research, we propose a hybrid model that combines machine-learning methods with Theta statistical method for more accurate prediction of future solar power generation from renewable energy plants. The machine learning models include long short-term memory (LSTM), gate recurrent unit (GRU), AutoEncoder LSTM (Auto-LSTM) and a newly proposed Auto-GRU. To enhance the accuracy of the proposed Machine learning and Statistical Hybrid Model (MLSHM), we employ two diversity techniques, i.e. structural diversity and data diversity. To combine the prediction of the ensemble members in the proposed MLSHM, we exploit four combining methods: simple averaging approach, weighted averaging using linear approach and using non-linear approach, and combination through variance using inverse approach. The proposed MLSHM scheme was validated on two real-time series datasets, that sre Shagaya in Kuwait and Cocoa in the USA. The experiments show that the proposed MLSHM, using all the combination methods, achieved higher accuracy compared to the prediction of the traditional individual models. Results demonstrate that a hybrid model combining machine-learning methods with statistical method outperformed a hybrid model that only combines machine-learning models without statistical method.

The main scope of the paper is to demonstrate the capabilities of PYTHIA forecasting platform, to compare time series forecasting techniques, which were used to forecast mortgage loans in UK, and to show how PYTHIA can be useful for a bank.

The paper outlines the methods used to forecast the time series data, which are included in PYTHIA. Theta, the time‐series used to forecast average mortgage loan prices, were grouped in: all buyers – average loan prices in UK; first‐time buyers – average loan prices in UK; and home‐movers – average loan prices in UK. The case of all buyers – average loan prices in UK, was presented in detail.

After the comparison of the methods, the best forecasts are produced by WINTERS and this is maybe due to the fact that there is seasonality in the data. The Theta method comes next in the row and generally produces good forecasts with small mean absolute percentage errors. In order to tell with grater certainty which method produces the most accurate forecasts we could compare the rest error statistics provided by PYTHIA too.

The paper presents the PYTHIA forecasting platform and shows how it can be used by the managers of a Bank to forecast mortgage loan values. PYTHIA can provide the forecasts required by practically all business situations demanding accurate predictions. It is designed and developed with the purpose of making the task of managerial forecasting straightforward, user‐friendly and practical. It incorporates a lot of knowledge and experience in the field of forecasting, modeling and monitoring while fully utilizing new capabilities of computers and software.

Angular errors in the robot axes can make a significant contribution to robot positioning accuracy. This paper seeks to propose a new measuring method for measuring angular errors.

New techniques were devised for the detailed investigation of joint angular errors using a reference encoder together with a precision electronic level and autocollimator. This equipment enabled vertical and horizontally orientated joint axes to be measured with the robot located on‐site. Circle contouring measurements were also undertaken to assess the significance of multi‐axis movements on the accuracy of the end effector.

The technique, devised using a simulation program for the robot geometry with results from a circular test, enables robot errors to be characterised in terms of datum location error, backlash, gear transmission error, axes misalignments and joint encoder offset.

The paper describes the experimental and theoretical accuracy characteristics of an articulated industrial robot. Close correlation was obtained between the experimental and theoretical results. This paper offers the practical robot calibration method for industrial application.

Large deformation problems are frequently encountered in various fields of geotechnical engineering. The particle finite element method (PFEM) has been proven to be a promising method to solve large deformation problems. This study aims to develop a computational framework for modelling the hydro-mechanical coupled porous media at large deformation based on the PFEM.

The PFEM is extended by adopting the linear and quadratic triangular elements for pore water pressure and displacements. A six-node triangular element is used for modelling two-dimensional problems instead of the low-order three-node triangular element. Thus, the numerical instability induced by volumetric locking is avoided. The Modified Cam Clay (MCC) model is used to describe the elasto-plastic soil behaviour.

The proposed approach is used for analysing several consolidation problems. The numerical results have demonstrated that large deformation consolidation problems with the proposed approach can be accomplished without numerical difficulties and loss of accuracy. The coupled PFEM provides a stable and robust numerical tool in solving large deformation consolidation problems. It is demonstrated that the proposed approach is intrinsically stable.

The PFEM is extended to consider large deformation-coupled hydro-mechanical problem. PFEM is enhanced by using a six-node quadratic triangular element for displacement and this is coupled with a four-node quadrilateral element for modelling excess pore pressure.

This paper aims to present a computationally efficient finite element model for the simulation of isothermal immiscible two-phase flow in a rigid porous media with a particular application to CO₂ sequestration in underground formations. Focus is placed on developing a numerical procedure, which is effectively mesh-independent and suitable to problems at regional scales.

The averaging theory is utilized to describe the governing equations of the involved unsaturated multiphase flow. The level-set (LS) method and the extended finite element method (XFEM) are utilized to simulate flow of the CO₂ plume. The LS is employed to trace the plume front. A streamline upwind Petrov-Galerkin method is adopted to stabilize possible occurrence of spurious oscillations due to advection. The XFEM is utilized to model the high gradient in the saturation field front, where the LS function is used for enhancing the weighting and the shape functions.

The capability of the proposed model and its features are evaluated by numerical examples, demonstrating its accuracy, stability and convergence, as well as its advantages over standard and upwind techniques. The study showed that a good combination between a mathematical model and a numerical model enables the simulation of complicated processes occurring in complicated and large geometry using minimal computational efforts.

A new computational model for two-phase flow in porous media is introduced with basic requirements for accuracy, stability, and convergence, which are met using relatively coarse meshes.

This study aims to investigate the impact of perceived values (hedonic and utilitarian), trust and subjective norms on consumers' purchasing intentions of organic food in Saudi Arabia; it also explores the moderating influence of availability on the relationship between the intentions of consumers and their actual purchasing behaviour.

A survey with 236 consumers of organic food in Saudi Arabia was carried out. The convergent and discriminant validity of latent variables was confirmed. The relationships among them were tested using Partial Least Square Modelling (PLS).

The results indicate that utilitarian and hedonic values, trust and subjective norms positively affect consumer purchase intention. They also reveal the moderating effect of availability on the relationship between consumers' purchasing intention and their actual behaviour in the Saudi Arabian context.

The study contributes to knowledge about the relationships among perceived values, trust, subjective norms, availability and consumer purchasing intentions of organic food, and their actual behaviour in an emerging market. The results enlarge the understanding of consumers' purchasing behaviour in the Saudi Arabian organic food market and point out some opportunities for future research.

The study is original in investigating the factors that influence customers' intention and their actual purchasing behaviour toward organic food in Saudi Arabia. It is a first attempt to test the moderating influence of availability on the relationship between purchase intention and actual purchasing behaviour toward organic food products in an emerging market.

The effects of normal surface suction and blowing on the Strouhal frequencies in vortex shedding over porous square cylinders was analysed numerically. The general characteristics determined were (1) an initial increase followed by a decreasing behaviour in the Strouhal frequency with increasing suction velocity and (2) a decrease in the Strouhal frequency with increasing blowing velocity. The numerical results were compared to an existing preliminary model, yielding fairly close agreement.

The purpose of this paper is to describe a real‐world system developed for a large food distribution company which requires forecasting demand for thousands of products across multiple warehouses. The number of different time series that the system must model and predict is on the order of 10⁵. The study details the system's forecasting algorithm which efficiently handles several difficult requirements including the prediction of multiple time series, the need for a continuously self‐updating model, and the desire to automatically identify and analyze various time series characteristics such as seasonal spikes and unprecedented events.

The forecasting algorithm makes use of a hybrid model consisting of both statistical and heuristic techniques to fulfill these requirements and to satisfy a variety of business constraints/rules related to over‐ and under‐stocking.

The robustness of the system has been proven by its heavy and sustained use since being adopted in November 2009 by a company that serves 91 percent of the combined populations of Australia and New Zealand.

This paper provides a case study of a real‐world system that employs a novel hybrid model to forecast multiple time series in a non‐static environment. The value of the model lies in its ability to accurately capture and forecast a very large and constantly changing portfolio of time series efficiently and without human intervention.

The purpose of this paper was to investigate the response of good and poor sleepers toward audio-visual stimulation via prefrontal theta EEG measurement.

The experiment included ten healthy subjects that were chosen after going through the Pittsburgh Sleep Quality Index (PSQI). They were divided into two groups that include five good and five poor sleepers. Next, in order to clarify the effects of audio-visual biofeedback during daytime, each subject was asked to go through six two-minute tasks that include: pre-baseline, eyes open at rest, eyes closed at rest, audio biofeedback with eyes open, video biofeedback also with eyes open, and post-baseline.

In Task 4, the audio stimulation task, both types of sleepers elicited higher theta waves due to demand in mental activity and also a meditation state. It was significantly higher in poor sleeper that demonstrated a peak difference of 25 percent compared to its good sleeper counterpart. In Task 5, the visual stimulation task, through the use of random numbers having blue and red color background, the theta amplitudes of good and poor sleepers drop together, due to beta waves becoming dominant, as the task required attention and focussed accounting for reduced theta amplitudes. The study was able to prove the use of prefrontal EEG in measuring and evaluating sleep quality by examining theta variation.

This paper proposed a novel and convenient method for evaluating sleep quality by utilizing only a single channel prefrontal EEG measurement.