Search results

Agriculture is the backbone of a country, contributing more than half of the sector of economy throughout the world. The need for precision agriculture is essential in evaluating the conditions of the crops with the aim of determining the proper selection of pesticides. The conventional method of pest detection fails to be stable and provides limited accuracy in the prediction. This paper aims to propose an automatic pest detection module for the accurate detection of pests using the hybrid optimization controlled deep learning model.

The paper proposes an advanced pest detection strategy based on deep learning strategy through wireless sensor network (WSN) in the agricultural fields. Initially, the WSN consisting of number of nodes and a sink are clustered as number of clusters. Each cluster comprises a cluster head (CH) and a number of nodes, where the CH involves in the transfer of data to the sink node of the WSN and the CH is selected using the fractional ant bee colony optimization (FABC) algorithm. The routing process is executed using the protruder optimization algorithm that helps in the transfer of image data to the sink node through the optimal CH. The sink node acts as the data aggregator and the collection of image data thus obtained acts as the input database to be processed to find the type of pest in the agricultural field. The image data is pre-processed to remove the artifacts present in the image and the pre-processed image is then subjected to feature extraction process, through which the significant local directional pattern, local binary pattern, local optimal-oriented pattern (LOOP) and local ternary pattern (LTP) features are extracted. The extracted features are then fed to the deep-convolutional neural network (CNN) in such a way to detect the type of pests in the agricultural field. The weights of the deep-CNN are tuned optimally using the proposed MFGHO optimization algorithm that is developed with the combined characteristics of navigating search agents and the swarming search agents.

The analysis using insect identification from habitus image Database based on the performance metrics, such as accuracy, specificity and sensitivity, reveals the effectiveness of the proposed MFGHO-based deep-CNN in detecting the pests in crops. The analysis proves that the proposed classifier using the FABC+protruder optimization-based data aggregation strategy obtains an accuracy of 94.3482%, sensitivity of 93.3247% and the specificity of 94.5263%, which is high as compared to the existing methods.

The proposed MFGHO optimization-based deep-CNN is used for the detection of pest in the crop fields to ensure the better selection of proper cost-effective pesticides for the crop fields in such a way to increase the production. The proposed MFGHO algorithm is developed with the integrated characteristic features of navigating search agents and the swarming search agents in such a way to facilitate the optimal tuning of the hyperparameters in the deep-CNN classifier for the detection of pests in the crop fields.

This paper focusses on on-demand food delivery (ODFD), i.e. the delivery of freshly prepared meals to customers' homes, enabled by the use of online platforms. In ODFD, a key process is represented by last-mile deliveries (LMDs): they directly affect customers (the delivery price influences their purchase intention), riders (the compensation drives their willingness to perform deliveries) and platforms (deliveries are very expensive). In this context, this work aims to investigate the economic performances of ODFD LMDs.

This study adopts a multi-method threefold process. First, it develops a model that – after the generation of customers' demand and the assignment of deliveries to available riders – identifies incomes and costs faced by an ODFD operator. Second, the model is applied to a base case in Milan (Italy). Third, sensitivity analyses are performed (on daily demand and riders' salary).

The analyses allow – besides the identification of significant values associated to ODFD profitability – to draw general insights about delivery price (e.g. free delivery is not economically sustainable), daily demand (e.g. greater demand values do not only improve positive results but also worsen negative ones) and fixed/variable wage mix (e.g. increasing the variable wage enhances the profitability for platforms).

On the academic side, this word enhances extant literature about ODFD, proposing a model – with multidisciplinary implications – to strategically investigate profitability conditions of LMDs. On the managerial side, it provides support for (logistics/marketing) ODFD practitioners since it allows to evaluate the potential impact of significant decisions on profitability.

Drawing on the social identity theory, this paper explores the impact of rider–driver ethnicity match on the driver’s expected ride satisfaction and willingness to perform, and rider’s trust on the driver.

The study relies on scenario-based online experiments with 291 ride-hailing drivers and 282 riders in the USA.

The findings indicate that ethnicity match between ride-hailing drivers and riders positively impact driver’s ride satisfaction and willingness to perform, and rider’s trust in the driver. The study also revealed a significant positive moderation effect of ethnic identity on the relationship of ethnicity match and those constructs.

While it may be challenging to influence an individual’s level of ethnic identity, managers can take steps to educate and train their employees regarding the impact of ethnic identity and discrimination, with a particular focus on those individuals who possess a strong sense of ethnic identity.

The findings of this research provide theoretical contributions to the existing literature on ride-hailing services and adds to the limited stream of logistics research that examines the impact of ethnicity on ride-hailing operations.

As businesses become more sophisticated and welcome new technologies, artificial intelligence (AI)-based methods are increasingly being used for firms' pricing decisions. In this review article, we provide a survey of research in the area of AI and pricing. On the upside, research has shown that algorithms allow companies to achieve unprecedented advantages, including real-time response to demand and supply shocks, personalized pricing, and demand learning. However, recent research has uncovered unforeseen downsides to algorithmic pricing that are important for managers and policy-makers to consider.

The purpose of this paper is to develop a model‐based methodology for the repetitive testing of multiple products with limited capacity, when the testing process is imperfect.

In a repetitive testing process, items that are classified as non‐conforming may be conforming, resulting in excessive scrapping of good items. Failed items are commonly retested to reduce scrapping costs. This paper develops a stochastic optimization formulation and its solution to determine the numbers of repetitive tests for multiple products that minimize the sum of the expected scrapping costs and variable testing costs, subject to a testing equipment capacity constraint. It also develops a procedure to estimate the parameter values that are used in the optimization formulation.

Computational experiments are conducted to evaluate the estimation and solution procedure and to understand the effect of testing machine capacity on the optimal total cost. These results demonstrate the viability of the proposed approach and the criticality of accurate parameter estimation.

This research shows the usefulness of the proposed optimization/statistical estimation approach to a real‐life complex inspection problem. However, the proposed model has to be modified when the characteristics of the testing equipment are changed.

The authors capture the idiosyncrasies in semiconductor manufacturing such as the high outgoing quality level, the repetitive testing environment, the high coefficient of variation in the number of failure products, and the testing capacity constraint. Conducting extensive computational experiments, the authors demonstrate that the proposed approach is viable.

The paper describes a complex, real‐life inspection management situation, develops a rigorous model‐based solution approach, and carefully demonstrates its viability.

The study proposes an evaluation model that allows quantitative characterization of the effects of service quality on consumer’s satisfaction for online food delivery (OFD) services in a nonlinear manner. As such, the authors endeavor to bridge the research-to-practice gaps whereby the effect magnitudes and nonlinear patterns of service quality have been overlooked in the current literature.

The quantitative Kano method is adopted. A Kano questionnaire was first developed by synthesizing and operationalizing existing evidence on OFD service qualities. The questionnaire solicited consumers’ evaluations of 21 OFD service attributes, and it was distributed to an online panel in Singapore. With 580 valid responses, the functions that quantitatively depict effects of each attribute on consumer’s satisfaction were subsequently derived.

The results reveal that among Singaporean consumers, food quality, reliability of delivery, responsiveness of customer support, ease-of-use of digital interfaces and promotions are pivotal attributes contributing to above-average satisfaction improvement across all performance levels. Meanwhile, delivery riders’ attitudes and real-time tracking functions emerge as substantial contributors to satisfaction at high-performance levels.

The findings provide crucial insights for OFD practitioners in Singapore in resource prioritization and service optimization. This study demonstrated the importance of streamlining customer support services and focusing on the utilitarian aspects of OFD services. Moreover, these results can be employed in advanced service improvement procedures, providing a roadmap for future OFD service enhancements.

This study pioneers the development of a quantitative quality evaluation model in the OFD context. With the established quantitative Kano model, the study addresses the omission of effect magnitudes and nonlinear patterns of service quality. It highlights the transition from a binary “does it affect satisfaction” to a more nuanced “how much does it affect satisfaction” approach, offering a robust understanding of consumer’s satisfaction dynamics.

The purpose of this paper is to analyze surface deformations caused by shear and moment forces on tactile materials and present a method to detect and reduce the risk of slippage by controlling the normal force as measured by tactile sensor arrays.

A predictive model has been proposed which uses a basic method adapted to real applications in grasp optimization. Prevention of premature release with minimum prehension force is addressed without the need to measure the coefficient of friction between object and robot gripper. Predictive models have been used to develop a set of rules which predict the pre‐slip based on fluctuations in tactile signal data.

The tactile sensors can be used in a “nonlinear” manner during manipulation tasks. When the gripper finger first makes contact with an object, the stress distribution under the finger skin varies rapidly. Predictive models have been used to develop a set of rules which predict the pre‐slip based on fluctuations in tactile signal data. Pre‐slip at the contact area just prior to object movement produces rapid but detectable stress transients.

Tactile sensors do not measure stress generated by a contact with an object directly, but instead measure strain in an interposed compliant, polymeric medium intended for sensor protection and prehension assistance. Reliable detection of pre‐slip has hitherto eluded researchers using such tactile techniques.

The aim of this paper is to generate a streamlined, transparent and effective instrument to fairly measure the contribution made by each student to a group project within a higher education context. The primary aim is to moderate the grades of underperforming students at the end of the project. There is a secondary benefit in alerting underperforming students to raise their contribution mid-task or face a potentially reduced grade at the final stage.

The development of this multi-dimensional instrument is guided by findings from previous research. The quest is to minimise the instructor's administrative work load in applying a moderation-only instrument that is open-source and available at no cost. Based on the literature, the survey instrument seeks to apply a peer-based, equitable and transparent evaluation of each member's contribution to a group task. The survey is applied at mid-task and again at end-task in order to afford underperformers the opportunity to address contribution deficits during the final phase of the project.

The instrument, called TANDEM©, offers a transparent, streamlined, equitable, confidential and practical measure of each student's contribution to a graded group task. Students whose end-task contribution falls below the group average rating receive a proportional reduction in their personal grade. Additionally, the end-task moderation instrument captures a single-item holistic measure of relative contribution that may, in the future, serve as a surrogate for the multi-dimensional measures currently in place.

TANDEM© was developed with group sizes of four or five members in mind. There is no evidence to support its application to three-person groups. Moreover, the application was applied only amongst under-graduate students. It is yet to be applied across post-graduate groups and within online learning environments. Future research into diverse cultural settings would serve to advance understanding of how moderation is perceived across borders.

Several existing group grade moderation methods propose complex algorithms that are “black box” solutions from a student's perspective. In establishing a fair, streamlined, confidential and transparent process for peer-rated moderation, TANDEM© deploys a concise instrument with a relatively small administrative load. TANDEM © may be applied to all groups or can selectively be applied to groups that report moderate, strong or extreme levels of conflict.

Students will appreciate the opportunity to rate peer contributions to group projects. This will dissipate the negative social sentiment that may arise when fellow students benefit from the work of others. Those students seeking conflict resolution within the group will value the transparent and equitable moderation of grades as well as the positive social implications that follow.

This research forms part of an ongoing quest to present a moderation instrument that fairly identifies student contribution to a group project. Whilst the solution proposed is one of many existing alternatives, its focus is on a practical moderation-only instrument that can immediately be applied to a course or major. The benefits lie in the ease of application and minimal administrative workload. This constitutes an original contribution to the individual (course or major) coordinator who seeks to apply a moderation-only instrument without having to commit to an extensive, broad-based group optimisation programme.

Software-define vehicular networks (SDVN) assure the direct programmability for controlling the vehicles with improved accuracy and flexibility. In this research, the resource allocation strategy is focused on which the seek-and-destroy algorithm is implemented in the controller in such a way that an effective allocation of the resources is done based on the multi-objective function.

The purpose of this study is focuses on the resource allocation algorithm for the SDVN with the security analysis to analyse the effect of the attacks in the network. The genuine nodes in the network are granted access to the communication in the network, for which the factors such as trust, throughput, delay and packet delivery ratio are used and the algorithm used is Seek-and-Destroy optimization. Moreover, the optimal resource allocation is done using the same optimization in such a way that the network lifetime is extended.

The security analysis is undergoing in the research using the simulation of the attackers such as selective forwarding attacks, replay attacks, Sybil attacks and wormhole attacks that reveal that the replay attacks and the Sybil attacks are dangerous attacks and in future, there is a requirement for the security model, which ensures the protection against these attacks such that the network lifetime is extended for a prolonged communication. The achievement of the proposed method in the absence of the attacks is 84.8513% for the remaining nodal energy, 95.0535% for packet delivery ratio (PDR), 279.258 ms for transmission delay and 28.9572 kbps for throughput.

The seek-and-destroy algorithm is one of the swarm intelligence-based optimization designed based on the characteristics of the scroungers and defenders, which is completely novel in the area of optimizations. The diversification and intensification of the algorithm are perfectly balanced, leading to good convergence rates.