Minimizing the total completion time in a two-machine flowshop problem with time delays

This article proposes to solve the problem of minimizing the total completion time in a two-machine permutation flowshop environment in which time delays between the machines are considered. For this purpose, an enumeration algorithm based on the branch-and-bound framework is developed, which includes new lower and upper bounds as well as dominance rules. The computational study shows that problems with up to 40 jobs can be solved in a reasonable amount of time.     

Potential effects of ultrafiltration process and date powder on textural,sensory, bacterial viability,antioxidant properties and phenolic profile of dromedary Greek yogurt

The aim of this study was to improve textural and antioxidant capacity of dromedary yogurt using ultrafiltration process and date powder. Ultrafiltration increased total solids content of dromedary milk within the range considered optimal to develop yogurt. Texture profile of Greek yogurt fortified with date powder (GYD) improved considerably compared to control. Sample of GYD are more appealing to consumer than control. LC-ESI-MS analysis of GYD extracts allowed the identification of fifteen phenolic compounds, among which quinic acid and cirsiliol were found to be the major phenolic acid and flavonol, respectively. GYD exhibited the highest DPPH•-radical scavenging activity, Fe²⁺ chelating capacity and Fe³⁺ reducing power. The follow-up of physical and microbiological stability of GYD and control during cold storage showed that date powder significantly increased bacterial counts and decreased syneresis. Therefore, ultrafiltration and date powder could be valued as effective tool to solve the poor consistency of dromedary milk products.     

Adaptive random forests for evolving data stream classification

Random forests is currently one of the most used machine learning algorithms in the non-streaming (batch) setting. This preference is attributable to its high learning performance and low demands with respect to input preparation and hyper-parameter tuning. However, in the challenging context of evolving data streams, there is no random forests algorithm that can be considered state-of-the-art in comparison to bagging and boosting based algorithms. In this work, we present the adaptive random forest (ARF) algorithm for classification of evolving data streams. In contrast to previous attempts of replicating random forests for data stream learning, ARF includes an effective resampling method and adaptive operators that can cope with different types of concept drifts without complex optimizations for different data sets. We present experiments with a parallel implementation of ARF which has no degradation in terms of classification performance in comparison to a serial implementation, since trees and adaptive operators are independent from one another. Finally, we compare ARF with state-of-the-art algorithms in a traditional test-then-train evaluation and a novel delayed labelling evaluation, and show that ARF is accurate and uses a feasible amount of resources.     

Coupling Genetic Local Search and Recovering Beam Search algorithms for minimizing the total completion time in the single machine scheduling problem subject to release dates

In this paper we consider the well-known single machine scheduling problem with release dates and minimization of the total job completion time. For solving this problem, denoted by 1|r_j|∑C_j, we provide a new metaheuristic which is an extension of the so-called filtered beam search proposed by Ow and Morton [30]. This metaheuristic, referred to as a Genetic Recovering Beam Search (GRBS), takes advantages of a Genetic Local Search (GLS) algorithm and a Recovering Beam Search (RBS) in order to efficiently explore the solution space. In this paper we present the GRBS framework and its application to the 1|r_j|∑C_j problem. Computational results show that it consistently yields optimal or near-optimal solutions and that it provides interesting results by comparison to GLS and RBS algorithms. Moreover, these results highlight that the proposed algorithm outperforms the state-of-the-art heuristics.