A comparative study of neural network and Box-Jenkins ARIMA modeling in time series prediction

This paper aims to investigate suitable time series models for repairable system failure analysis. A comparative study of the Box-Jenkins autoregressive integrated moving average (ARIMA) models and the artificial neural network models in predicting failures are carried out. The neural network architectures evaluated are the multi-layer feed-forward network and the recurrent network. Simulation results on a set of compressor failures showed that in modeling the stochastic nature of reliability data, both the ARIMA and the recurrent neural network (RNN) models outperform the feed-forward model; in terms of lower predictive errors and higher percentage of correct reversal detection. However, both models perform better with short term forecasting. The effect of varying the damped feedback weights in the recurrent net is also investigated and it was found that RNN at the optimal weighting factor gives satisfactory performances compared to the ARIMA model.     

Examining the Effectiveness of Gamification in Human Computation

Within the human computation paradigm, gamification is increasingly gaining interest. This is because an enjoyable experience generated by game features can be a powerful approach to attract participants. Although potentially useful, little research has been conducted into understanding the effectiveness of gamification in human computation. In this experimental study, we operationalized effectiveness as perceived engagement and user acceptance and examined it by comparing the performance of a gamified human computation system against a non-gamified version. We also investigate the determinants of acceptance and how their effects differ between these two systems. Analysis of our data found that participants experienced more engagement and showed higher behavioral intentions toward the gamified system. Moreover, perceived output quality and perceived engagement were significant determinants of acceptance of the gamified system. In contrast, determinants for acceptance of the non-gamified system were perceived output quality and perceived usability.     

Screening of Chinese medicinal herbs for bioactivity against Sitophilus zeamais Motschulsky and Tribolium castaneum (Herbst)

Extracts of 40 species of Chinese medicinal herb from 32 different botanical families were screened for contact, fumigant and feeding-deterrent activities against the two stored-grain insects Sitophilus zeamais and Tribolium castaneum. Thirty Chinese medicinal herbs exhibited insecticidal or feeding-deterrent activities against the two species of insects. Extracts of Artemisia argyi, Dictamnus dasycarpus, Evodia rutaecarpa, Litsea cubeba, Narcissus tazetta var. chinensis, Polygonum aviculare, Rhododendron molle, Sophora flavescens, Stemona sessilifolia, Tripterygium wilfordii, and Torreya grandis were most active.     

Tool wear detection and fault diagnosis based on cutting force monitoring

In metal cutting processes, an effective monitoring system, which depends on a suitably developed scheme or set of algorithms can maintain machine tools in good condition and delay the occurrence of tool wear. In this paper, an approach is developed for fault detection and diagnosis based on an observer model of an uncertain linear system. A robust observer is designed, using the derived uncertain linear model, to yield the necessary and key information from the system. Subsequently, it is used as a state (tool wear) estimator, and fault detection is carried out by using the observed variables and cutting force. The developed approach is applied to milling machine center. Several linear models are identified based on different working conditions. A dominant model plus uncertain terms is derived from these model set and used as an observer. Threshold values are proposed for detecting the fault of the milling machine. Examples taken from experimental tests shown that the developed approach is effective for the fault detection. The approach can be used for fault detection of failures arising from sensor or actuator malfunction.     

Sequential Data Fusion via Vector Spaces: Fusion of Heterogeneous Data in the Complex Domain

A sequential data fusion approach via higher dimensional vector spaces is introduced. This is achieved by making use of the representation of directional signals within the field of complex numbers $C$ . The concept of data fusion is next introduced and the place of the proposed approach within that framework is identified. The benefits of such an approach are illustrated and a range of possible applications is shown. The concept introduced is supported by a real world case study which focuses on simultaneous forecasting of wind speed and direction. The architectures and learning algorithms which support this concept are introduced and their distributed sequential fusion nature is highlighted.     

Effects of urea fertigation of apple trees on soil pH,exchangeable cations and extractable manganese in a sandy loam soil in New Zealand

Changes in soil pH, exchangeable aluminium (Al), calcium (Ca), magnesium (Mg), and potassium (K) and extractable manganese (Mn) were investigated after urea fertigation of a sandy loam soil in an apple orchard in New Zealand. Urea at three rates (0, 25, 50 kg N ha^–1 yr^–1 or 0, 16.9, 33.8 g N emitter^–1 yr^–1) was applied in 4 equal fertigations. Soil cores at 4 profile depths (0–10, 10–20, 20–40 and 40–60 cm) directly below and 20 cm from the emitter were sampled approximately 4 weeks after each fertigation and in the following winter. Results obtained showed that the largest changes in soil pH and cations occurred in soils directly below the emitter in the 50 kg N ha^–1 yr^–1 treatment where the soil pH decreased by 1.6 pH units at all soil depths. The lowest pH of 4.3 was observed at a depth of 27 cm. Exchangeable Al and extractable Mn levels increased to 11 meq kg^–1 and 78µg g^–1 respectively. Estimated losses of Ca, Mg and K from the upper soil profile depth (0–10 cm) represented 23, 63 and 27% of their respective total exchangeable levels. At lower profile depths (>20 cm), accumulation of displaced K was evident. Variable, and generally non-significant, chemical changes recorded in soils 20 cm from the emitter were attributed to restricted lateral water movement, and therefore urea movement, down the profile.The present study showed that one season of urea fertigation by trickle emitters, applied to a sandy loam, at half the rate conventionally applied to apple orchards (50 kg N ha^–1 yr^–1) resulted in pH and mineral element imbalances which were potentially and sufficiently severe to inhibit tree growth.     

Characterisation of gold kiwifruit pectin isolated by enzymatic treatment

Pectin was extracted from gold kiwifruit by four commercial enzyme preparations (Celluclast 1.5 L, Cytolase CL, Cellulyve TR 400 and NS33048). The chosen enzymes were used either in single or in combination, with and without protease addition. The recovered pectin was characterised and compared for the yield, total nonstarch polysaccharide and neutral sugar composition, protein and ash, pectin, molecular weight distribution and the viscosity. Results indicated that enzyme‐extracted gold kiwifruit pectin was rich in galacturonic acid. Purified pectin yield, and their physicochemical composition and rheological property (viscosity), was significantly affected by the type of enzyme used. The pectin extracted by Celluclast 1.5 L demonstrated to be the most viscous and recorded the highest in molecular weight (M_w) (1.65 × 10⁶ g mol^–1) compared with the other extracts. The extract M_w and their distribution were discussed and related to their viscosity behaviour.     

Benefits valuation of potable water quality improvement in Malaysia: the case of Kajang Municipality

This article estimates the benefits of potable water quality improvements in Kajang Municipality in the state of Selangor, Malaysia, using the avertive cost method. Households were willing to pay MYR 322 annually to improve potable water quality. This represents about 80% of their average annual water bill. The present value of aggregate benefits over a period of 30 years at various social discount rates ranged from MYR 301 million to MYR 768 million. These estimates can be used as a reference for public investment criteria. The findings suggest that there is merit in an upward revision of water tariffs if they provide improvements in water quality.