Abstract

With the continuous development of technology, traditional marketing methods no longer meet the needs of the main forces of social consumption, and people urgently need more innovative and personalized marketing strategies. E-commerce companies must develop a comprehensive customer-oriented marketing strategy based on big data and multi-channel to achieve their long-term healthy development. This paper first investigated the impact of the digital economy on e-commerce enterprises, focused on the transformation of the digital economy on the marketing model, expounded the development analysis of e-commerce in the digital economy era, and described the development trend of e-commerce marketing in the digital economy era. Then, this paper expounded the current problems faced by e-commerce enterprises, and discussed the lack of integrity, homogeneity, large-scale marketing strategies, and the lack of analysis and application of big data. After that, this paper put forward the marketing strategy of e-commerce enterprises in the digital economy era, and studied it from three aspects, namely, building a reasonable product management structure, marketing strategy based on customized marketing content, and social media marketing strategy based on information sharing. Then this paper proposed to use genetic algorithm to strengthen the marketing strategy of e-commerce enterprises. Finally, based on experiments and surveys, this paper used genetic algorithms to strengthen the construction of e-commerce enterprise marketing strategy in the digital economy, and concluded that the new e-commerce enterprise marketing strategy was 21% more satisfactory than the traditional new e-commerce enterprise marketing strategy. Through comparison, it can see that the integrity of the marketing plan of the new e-commerce enterprise’s marketing strategy was 0.33 higher than that of the traditional e-commerce enterprise’s marketing strategy, and the integrity of the promotion strategy was 0.34 higher than that of the traditional e-commerce enterprise’s marketing strategy. After using the new e-commerce enterprise marketing strategy, the improved management structure was 0.29 higher than that of the traditional monitoring system, and the high quality of products was 0.18 higher than that of the traditional system.

Introduction

Electronic commerce is the combination of Internet information technology and traditional offline business activities. With the development and maturity of Internet technology, e-commerce continues to grow. Its concept, connotation and meaning evolve gradually with the evolution of technology and actual e-commerce activities. From a historical perspective and combined with the development of information technology, the evolution of the concept of e-commerce can provide a new perspective for observing the development of new forms of e-commerce. In 2023, the Ministry of Commerce implemented the decisions and deployments of the Party Central Committee and the State Council to accelerate the development of the digital economy, promote the important role of e-commerce in restoring and expanding consumption, promoting the integration of data and reality, and deepening international cooperation, and achieve positive results in high-quality development. With the rapid development of the digital economy, e-commerce enterprises have also experienced a golden age of development. Through the analysis and research of relevant industry data, it is not difficult to find that e-commerce business cannot be separated from a large number of consumers. E-commerce enterprises can develop normally with enough consumers. In order to survive in the highly competitive market, e-commerce enterprises must make full use of advanced big data analysis technology and formulate marketing strategies suitable for the development of e-commerce enterprises. Main contribution: This article mainly focuses on the research of marketing strategies related to e-commerce enterprises in the context of big data, and proposes corresponding improvement strategies. Objective: This article aims to promote the healthy and long-term development of e-commerce enterprises in the context of big data through improved marketing strategies based on this foundation.

Conclusions

Currently, how to promote their own development through effective marketing strategies has become an important issue that e-commerce enterprises need to pay attention to in the context of big data. This article mainly studies the marketing strategy related issues of e-commerce enterprises with big data as the background, and proposes corresponding improvement strategies. This article aims to promote the healthy and long-term development of e-commerce enterprises in the context of big data through improved marketing strategies based on this foundation. To sum up, in the new economic era, the marketing strategy of enterprises is very important. The fierce competition between enterprises requires enterprises to constantly innovate, change marketing strategies and maximize market space. Therefore, every enterprise should strive to implement marketing strategies in an innovative spirit. Only by constantly changing marketing strategies and actively adapting to the needs of economic development can enterprises stand out from the competition. E-commerce companies are shifting from single and large-scale marketing strategies to homogeneous and diversified marketing strategies. E-commerce companies need to establish accurate integrated marketing based on big data, customer-oriented and multi-channel to survive and develop in the future competition. The development of genetic algorithms has led to exponential growth in the direct to consumer personal genetic testing industry (such as 23andMe, Ancestry, Microgenes, 23Rubik’s Cube, etc.), resulting in the formation of a vast private genetic database. The author explores the potential impact of genetic algorithms on the field of e-commerce. Based on the results of behavioral GA research, the author proposes to incorporate the influence of genetic factors into the existing theoretical research framework of e-commerce, and use this to examine the potential market applications of genetic algorithms. Genetic research design can be used to test causal relationships, and consumer theory can be improved by revealing the biological mechanisms of behavior. In the future, the author will further evaluate the moral and ethical challenges related to corporate autonomy, privacy, and discrimination, and propose a future research agenda.

Abstract

     Vehicular Ad Hoc Networks (VANET) is an emerging technology that enables a comfortable, safe, and efficient travel experience by providing mechanisms to execute applications related to traffic congestions, road accidents, autonomous driving, and entertainment. The mobile vehicles in VANET are characterized by low computational and storage capabilities. In such scenarios, to meet applications’ performance requirements, requests from vehicles are offloaded to edge and cloud servers. The high energy consumption of these servers increases operating costs and threatens the environment. Energy-aware offloading strategies have been introduced to tackle this problem. Existing works on computation offloading focus on optimizing the energy consumption of either the IoT devices/mobile/vehicles and/or the edge servers. This paper proposes a novel offloading algorithm that optimizes the energy of edge–cloud integrated computing platforms based on Evolutionary Genetic Algorithm (EGA) while maintaining applications’ Service Level Agreement (SLA). The proposed algorithm employs an adaptive penalty function to incorporate the optimization constraints within EGA. Comparative analysis and numerical experiments are carried out between the proposed algorithm, random and genetic algorithm-based offloading, and no offloading baseline approaches.

Introduction

     Vehicular Ad Hoc Networks (VANET) [1] is an emerging technology where vehicles acting as network nodes are equipped with computational resources and connectivity such as vehicle-to-vehicle (V2V), vehicle-to-infrastructure (V2I), vehicle-to-roadside (V2R), vehicle-to-sensors (V2S), vehicle-to-pedestrian (V2P), and vehicle-to-everything (V2X) communications. It enables a comfortable, safer, convenient, and efficient travel experience for users by using applications such as sending alerts for congestions and accidents, autonomous driving, video-enabled real-time navigation, interactive gaming, and entertainment [2], [3], [4]. These applications often require high computation and storage resources, and low latency to process complex operations. However, mobile vehicles have limited onboard computing and storage capabilities to process resource-intensive applications while maintaining the Quality of Services (QoS). To address this issue, a cloud-based vehicular network has been introduced. Cloud computing [5], [6] provides on-demand computational and storage resources to mobile vehicles over the Internet. The remote cloud servers have high computation capabilities that would satisfy applications processing times. However, a high latency between the vehicle and cloud resources hinders the deployment of time-critical applications such as autonomous driving. In addition, a delayed response for applications such as traffic congestion and interactive gaming becomes less reliable. To overcome vehicles-to-cloud latency issues, Vehicular Edge Computing (VEC) [7] has been introduced.

Conclusions and future work

     Computation offloading is important in edge–cloud integrated vehicular networks to execute computationally intensive applications having strict SLA requirements. However, the energy consumption of the edge–cloud integrated computing platform should be considered energy-efficiency is crucial. In this paper, an Energy-SLA-Aware evolutionary genetic algorithm is proposed for edge–cloud computation offloading in a vehicular network that executes a vehicle’s request either on the edge server to which the request is submitted or offloads the request to one of the cloud servers. The offloading decision is made in a way that the total energy consumption of a set of requests is minimized and the SLA requirements of each request are maintained in terms of latency and processing time. The SLA constraints in the proposed algorithm are handled using the adaptive penalty function. This is the first work to propose an energy-SLA-aware offloading in the vehicular network using EGA that optimizes the energy consumption of the edge and cloud servers simultaneously, while adhering to the latency and processing time constraints. Comparative analysis and numerical experiments carried out revealed that the proposed algorithm outperforms no offloading and random offloading approaches in terms of energy consumption, and no offloading, random and energy-non-SLA-aware genetic-based baseline approaches in terms of percentage of SLA violations.

1- Introduction

2- Related work

3- Research objectives

4- Distance based formation using GA

5- Discussion

6- Conclusions

References

Abstract

In this paper an alternative method to achieve distance based formation is presented. The method uses Genetic Algorithms to find a suitable solution based on angle and distance, and an appropriate constant velocity to avoid collisions. The designed algorithm is extended to a parallel scheme to improve its performance and achieve Artificial Distributed Intelligence, in which the robots share, through solution migration, the best ways to converge to desired distances while avoiding collisions, finally reaching consensus on the solution. The algorithm is tested using simulations and real robots experiments.

Introduction

Researchers of the fast growing field of Multi robot systems (MRS) focus on the idea that there is strength in numbers, as presented in Dudek’s example [1] of a task that requires simultaneous operation where the restrictions of a single robot system is overcome by a MRS. MRS made out of a large number of simple and cheap robots could have better performance than individual highly specialized and costly robots, mainly because: with more robots exploration of a wider spatial area can be faster and distributed specialized jobs can be carry out by different team members if the MRS is heterogeneous. Another important characteristic of MRS is redundancy, a large number of robots support each other carrying out teammates task if the robots fails to achieve the mission [2]. There is a lot of research in MRS on designing appropriate coordination strategies and control laws that enable them to achieve objectives efficiently together [3–5]. Coordination strategies include various problems that aim to drive a group of agents to some common state, this is usually called consensus, agreement, synchronization or rendezvous, in this paper this strategies are referred as the consensus problem, extensively studied in [6,7], modelling robots with single integrator and using graph theory describe topology, ergo, connections between robots. Control laws are designed to achieve consensus avoiding inter robot interference [8]. Global convergence, group stability, group following and other characteristics are studied through their topology graphs [2].

1- Introduction

2- The Principle of GA

3- The GALA software system

4- Conclusions

References

Abstract

This paper presents an Intelligent Information System for education. The system was created for teaching students to use genetic algorithm in application to optimization tasks. The system allows to quickly encode a solution of the problem and pick up most suitable configuration of genetic algorithm. The paper also demonstrates a specific example of usage of educational system to solve an optimization task. The paper contains a description of the educational system and its features, a description of its capabilities of working with genetic algorithm and its graphical interface.

Introduction

The use of information technologies (IT) in various spheres of human activity, the exponential growth of information volumes and the need to respond quickly in any situations necessitate the search for adequate ways to solve new and new emerging problems [1]. One of the most effective ways is the study, application, development and further improvement of algorithms and methods of machine computing. Finding the optimal solution is one of the problems that researchers face, both in the theoretical field and in solving engineering and production problems. Optimization problems imply the need to choose from the set of possible solutions the best from the point of view of certain criteria, satisfying given conditions and limitations.

This paper addresses a multi-objective feature selection problem for early time series classification. Previous research has focused on how many features to consider for a classifier, but has not considered the starting time of classification, which is also important for early classification. Motivated by this, we developed a mathematical model for which the objectives are to maximize classification performance and minimize the starting time and execution time of classification. We designed an efficient genetic algorithm to generate solutions with high probability. In experiment, we compared the proposed algorithm and general genetic algorithm under various experimental settings. From the experiment, we verified that the proposed algorithm can find a better feature set in terms of classification performance, starting time and execution time of classification than feature set found by general genetic algorithm.

Introduction

Time series classification is used to predict the class label of a time series instance by a welltrained classifier (Deng et al., 2013). That is, if a time series instance , ? ? (?) = (? , is given, its label, , is predicted by the classifier as (?) 1 , ? (?) 2 , ⋯, ? (?) ? ) ? (?) ?( ∙ ) ? (?) = ? (? . Time series classification is used to accomplish tasks in many fields, including fault (?) ) detection in the manufacturing field (Lee et al., 2017), disease diagnosis in the medical field (Lacy et al., 2018), and stock trend analysis in the financial field (Moews et al., 2019). Various classifiers such as neural networks (NNs) and support vector machines (SVMs) are employed and modified to classify time series. Ignatov (2018) employed a convolutional neural network (CNN) to recognize human activity from accelerometer data. Kim and Cho (2018) developed a C-LSTM (CNN- Long Short-Term Memory model) NN to detect anomalies in web traffic data. CNNs and LSTMs in the developed model extracted spatial features and temporal characteristics, respectively. Emoto et al. (2018) used NNs to detect low-intensity snoring episodes from a sleeping sound dataset. Cheng and Dong (2019) employed SVM technology to monitor the nanomachining process with respect to the machining performance. Kalantarian et al. (2016) used SVMs to segment streaming timeseries audio signals probabilistically.

1. Introduction

2. Modelling of steel-jacketing action and application to RC fiber-section columns

3. Design optimization framework

4. Case study structure and specific design optimization assumptions

5. Results of the optimization

6. Effectiveness of the optimization framework in case of structural irregularities

7. Conclusions

References

Abstract

Retrofitting of existing reinforced concrete (RC) frame structures by steel angles and battens (steel-jacketing) is a commonly employed technique used to retrofit beams and columns against gravity and seismic loads. Steeljacketing (SJ) effectively provides additional deformation and strength capacity to RC members but its application is associated with noticeable downtime of the building and non-negligible costs, depending on the amount of structural and non-structural manufacturing and materials. This paper presents an optimization framework aimed at the minimization of seismic retrofitting-related costs by an optimal placement (topological optimization) and amount of steel-jacketing reinforcement. In the proposed framework a 3D RC frame fiber-section model implemented in OpenSees is handled by a genetic algorithm routine that iterates reinforcement configurations to match the optimal solution. The feasibility of each solution is controlled by the outcomes of a static pushover analysis in the framework of N2 method. Results will provide optimized location and amount of steel-jacketing reinforcement, showing how effective and sustainable reduction of retrofitting costs is achievable maintaining a specified safety level.

Introduction

Retrofitting of reinforced concrete columns with cages arranged by steel angles and battens (steel jacketing) is a widely employed technique to improve strength and deformation capacity of beams and columns of existing buildings presenting critical conditions with respect to seismic and gravity loads. Steel jacketing of columns can be generally arranged in two ways. The first provides a moment resisting connection between the steel cages and the slabs (Fig. 1a). In this case, besides the confinement action exerted by the cage, additional flexural strength is provided. Since moment resisting connection are not always easy to realize, steel jacketing is often arranged by simply applying the cages (Fig. 1b). Even in this case, a certain additional flexural resistance is observed because of friction forces transfer between the steel angles and the concrete column (Campione et al. 2017 [1]), but the most significant contribution is related to the increase of deformation capacity as consequence of the strong confinement action. Experimental and numerical investigation have been carried out in the last years both for the first [2–5] and the second [1,6–10] typology of arrangement.

Despite its effectiveness in providing additional strength and deformation capacity to RC members, it should be said that steel jacketing is an invasive strengthening technique. In fact, the reinforcement of columns provides also the demolition and reconstitution of eventual portions of masonry infills and plaster. This is associated with significant direct costs and noticeable downtime for the building. A second issue regards the design of the intervention in terms of individuation of the columns to retrofit and the choice of the battens area and spacing. In fact, when approaching by non-linear static analysis (pushover), as a method to assess the performance before and after retrofitting intervention, a significant number of attempt iterations are needed to individuate the most suitable retrofitting configuration, especially when the number of columns is large and the building has irregular configuration. In the absence of a specific optimization process this generally brings the designer to adopt overall compromise solutions which allow obtaining effective seismic performance without optimization of the costs. Structural optimization is widely recognized as a valuable computational tool allowing engineers to obtain cost-effective designs.

The third party misuse and manipulation of digital images is a threat to security and privacy of human subjects. The requirements to fulfil the security needs of digital images have led to the development of good encryption techniques. The proposed method of encryption scheme combines the techniques of chaotic image encryption and DNA (Deoxyribonucleic Acid) sequence operations. The encryption mechanism is applied separately on three channels (R, G and B) of the colour image with a suitable chaotic map. The map selection algorithm selects a good chaotic map based on image properties and other parameters. A method of DNA encoding is also applied after chaotic encryption. A new Mandelbrot Set based conditional shift algorithm is introduced to apply confusion effectively on R, G and B channels. Simulation and security analysis confirm that the proposed image encryption scheme demonstrates extraordinary performance in terms of security.

Introduction

The fast development of digital technologies and communication networks causes to generate more and more digital data [1]. Security and privacy of the images captured from electronic devices are of paramount importance with the increase in velocity, volume and veracity of the acquired multi-modal data. One effective well-known method to keep integrity is to encrypt the images by turning it from a meaningful version to a meaningless one [2]. Images have large data capacities, and it is known that a high correlation between adjacent pixels. So the traditional cryptosystems, such as RSA (Rivest–Shamir–Adleman), DES (Data Encryption Standard), AES (Advanced Encryption Standard), are not suitable for image encryption [3]. Recently, different kinds of image encryption algorithms based on optical transform, chaotic systems, Fourier transform, cellular automata, wavelet transform, magic cube, etc. have been studied by groups of researchers from all over the world [4]. There exist two groups in image encryption techniques based on the method applied to design the encryption and decryption system. Chaotic cryptography is the application of the mathematical chaos theory (that is, the chaotic sequences produced by chaotic maps) in the generation of cryptographic algorithms.

1- Introduction

2- Backgrounds

3- Incremental genetic algorithm

4- Genetic framework

5- Experimental results

6- Conclusion

References

Abstract

Graph pattern matching is a key problem in many applications which data is represented in the form of a graph, and this problem is generally defined as a subgraph isomorphism. In this paper, we analyze an incremental hybrid genetic algorithm for the subgraph isomorphism problem considering various design issues to improve the performance of the algorithm. An incremental hybrid genetic algorithm was previously suggested to solve the subgraph isomorphism problem and have shown good performance. It decomposes the problem into a sequence of consecutive subproblems which has an optimal substructure. Each subproblem is solved by the hybrid genetic algorithm and the solutions obtained are extended to be applied to the next subproblem as initial solutions. We examine a wide range of schemes that determine the overall performance of the incremental process and make a number of experiments to verify the effectiveness of each scheme with the synthetic dataset of random graphs. We show that the performance of incremental approach can be significantly improved compared to the previous representative studies by applying appropriate schemes found by the experiments. In addition, we also investigate the effect of different genetic parameters and identify the scalability of our method by conducting experiments using real world dataset with large-sized graphs.

Introduction

Graph is a simple and universal data representation to model pairwise relationships among a set of objects. One of the interesting problems encountered when handling graph data is a graph pattern matching arising from pattern recognition, knowledge discovery, biology, cheminformatics, dynamic network traffic and intelligence analysis [1–4]. And this matching is typically formulated in terms of the subgraph isomorphism problem. Given two graphs G and H, the subgraph isomorphism problem is to determine whether H contains a subgraph that is isomorphic to G and this decision problem is well-known NP-Complete [5]. Many algorithms have been proposed to solve this problem starting with the backtracking algorithm by Ullmann [6]. VF2 [7], QuickSI [8], GraphQL [9], GADDI [10] and SPath [11] improved the performance by exploiting different join orders, pruning rules, and auxiliary information from the Ullmann algorithm [12]. The maximum common subproblem, the generalized problem of the subgraph isomorphism problem, also has been tackled by many algorithms [13–15]. However, these algorithms for both problems have exponential time complexity, their scalability are limited and they only work with auxiliary information such as vertex or edge labels. On the other hand, metaheuristic algorithms, especially a genetic algorithm, have been used to address this problem [16–22]. They can usually find good quality solutions within a reasonable amount of time, but most algorithms does not have enough search capability to cover large and complex problem space of the subgraph isomorphism problem.

I- Introduction

II- Genetic Algorithm

III- Travelling Salesman Problem

IV- Implementation

V- Results and Analysis Authors Figures

VI- Conclusion And Implication For Future Work

References

Abstract

Optimization problem is which mainly focuses on finding feasible solution out of all possible solutions. Travelling salesman problem belongs to this one. As it is not possible to find its solution in definite polynomial time that is why it is considered as one of the NP-hard problem. This paper utilizes the optimization capability of genetic algorithm to find the feasible solution for TSP. The algorithm starts with the calculation of Euclidean distance between the towns to be visited by the salesman. Initial chromosome pool is generated using value encoding. Then best fit chromosomes are selected by applying roulette wheel selection which then goes through m-point crossover. Now we apply interchange mutation on the offsprings generated before. Now this whole process is repeated until the convergence of genetic algorithm.

CONCLUSION AND IMPLICATION FOR FUTURE WORK

Heuristic methods and genetic approaches are the most appropriate ways to solve the travelling salesman problem. Multiple optimal solutions can be obtained for this problem by using various combinations of selection, crossover and mutation techniques. We have surveyed many approaches and many combinations of genetic operators for this problem and the used combination of genetic operators in this paper is the optimized one among them. For future extension of this work we can use multiple combinations of hybrid genetic operators. The approach used can be operated in diverse network optimization problems like vehicle navigation routing model, task scheduling models, Chinese postman problem and logistic networks.

1- Introduction

2- Related work

3- Problem definition

4- A Pareto optimal genetic algorithm for the student–supervisor allocation problem

5- Experiments

6- Conclusions

References

Abstract

The problem of allocating students to supervisors for the development of a personal project or a dissertation is a crucial activity in the higher education environment, as it enables students to get feedback on their work from an expert and improve their personal, academic, and professional abilities. In this article, we propose a multi-objective and near Pareto optimal genetic algorithm for the allocation of students to supervisors. The allocation takes into consideration the students and supervisors’ preferences on research/project topics, the lower and upper supervision quotas of supervisors, as well as the workload balance amongst supervisors. We introduce novel mutation and crossover operators for the student–supervisor allocation problem. The experiments carried out show that the components of the genetic algorithm are more apt for the problem than classic components, and that the genetic algorithm is capable of producing allocations that are near Pareto optimal in a reasonable time.

Introduction

Every year in higher education (HE) institutions, students undertake individual projects that are supervised by a tutor that offers academic advice and guidance, either as an undergraduate or master dissertation, as part of their coursework, or simply as a summer research project. Students are usually allocated to supervisors for their projects by means of a centralized human decision maker or by means of interactions between students and staff members. The decision makers have to take into consideration the preferences of both students and supervisors with respect to the conduct of the project, as well as departmental constraints such as minimum and maximum levels of workload (in terms of supervision) for each supervisor. This situation results in an extremely time consuming process, and a suboptimal allocation due to a large and complex search space faced by human decision makers. Automating this process by applying artificial intelligence techniques may enhance the process in terms of satisfaction and performance of students with these individual projects. In this article, we present a genetic algorithm for matching students to supervisors according to both students’ and supervisors’ preferences and the constraints of the department. The rationale behind this problem is matching an appropriate student with a supervisor for the development of an individual project.

Introduction

Description of model

Results and discussion

Conclusion

References

Abstract

Flexible manufacturing system (FMS) readily addresses the dynamic needs of the customers in terms of variety and quality. At present, there is a need to produce a wide range of quality products in limited time span. On-time delivery of customers’ orders is critical in make-to-order (MTO) manufacturing systems. The completion time of the orders depends on several factors including arrival rate, variability, and batch size, to name a few. Among those, batch size is a significant construct for effective scheduling of an FMS, as it directly affects completion time. On the other hand, constant batch size makes MTO less responsive to customers’ demands. In this paper, an FMS scheduling problem with n jobs and m machines is studied to minimize lateness in meeting due dates, with focus on the impact of batch size. The effect of batch size on completion time of the orders is investigated under following strategies: (1) constant batch size, (2) minimum part set, and (3) optimal batch size. A mathematical model is developed to optimize batch size considering completion time, lateness penalties and setup times. Scheduling of an FMS is not only a combinatorial optimization problem but also NP-hard problem. Suitable solutions of such problems through exact methods are difficult. Hence, a meta-heuristic Genetic algorithm is used to optimize scheduling of the FMS.

Introduction

An FMS is a highly automated production system consisting of a group of computer numerical control machine tools, linked by material handling system and controlled by a distributed control system. An FMS can simultaneously process medium variety and medium size volumes of products (Browne et al. 1984). An FMS is a sophisticated production system to respond dynamic variations of contemporary market including lead time reduction, flexibility to respond market variations and higher productivity (Atmani and Lashkari 1998). The major problems that an FMS has to face include designing, planning, scheduling and controlling. Among these, scheduling problem is a major challenge (Stecke 1985). An FMS’s scheduling differs from a conventional job shop scheduling due to routing flexibility of parts (Jain and Elmaraghy 1997). Scheduling is the assignment of resources over time to perform tasks. In conventional scheduling system, only one resource is considered, which processes the parts. An FMS scheduling differs from conventional job shop scheduling. The complexity of FMS scheduling is due to the flexibility of an FMS in terms of machine, product, operation and routing (Browne et al. 1984). Scheduling of an FMS is effected by several factors including orders arrivals, due dates and batch size (Liu and MacCarthy 1996). The production management has to set policies whether to handle periodic or continuous orders of the customers. Meeting customers’ due dates play key role in make to order system and the FMSs are mostly suited for MTO system.

Task offloading plays a critical role in cloud networked multi-robot systems for leveraging computation support from cloud infrastructure and benefiting greatly from the well-developed cloud network facilities. However, considering the delay constraint, the extra costs of data transmission and remote computation, it is not trivial to make optimized offloading decisions. In particular, task offloading for robots is more complex due to their on-demand mobility and network connectivity that significantly influence the robot–cloud communication links. Moreover, for multi-robot systems, a suitable balance of workload between local network (robot–robot) and global cloud (robot–cloud) is also required, so as to attain proper utilization of resources. Therefore, it is essential to establish more comprehensive offloading schemes for modeling systems that can handle these higher level of complications. With that view, this paper aims to develop a novel multi-layer decision-making scheme for task offloading which jointly considers the following four aspects: (i) selection of task for offloading, (ii) selection of robot to offload a task, (iii) selection of location to offload/perform task, (iv) selection of access point for offloaded task. An integrated framework for cloud networked multi-robot systems is presented to enable our task offloading scheme where the primary robot can aid from additional local robots to improve the offloading process. In particular, we consider a warehouse scenario with 36 cell workspace where a 40 node taskflow is motivated from a “parcel sorting and distribution” application, to be completed by the primary robot. The offloading decision for each task is formulated as part of a joint optimization problem and it is solved by developing a multi-layer genetic algorithm scheme that takes into account motion, network connectivity and local sharing for its offloading decisions. We evaluate the results of the scheme via comparison with two validated benchmarks. The outcome highlights a significant improvement in overall system performance due to joint involvement of motion (path planning), connectivity (bandwidth estimation) and robot–robot communication (local offloading) that facilitates energy-efficient offloading to cloud, faster completion of tasks and better utilization of available resources.

Introduction

The widespread innovation in robotic technology and subsequent increase in their computing capabilities are enabling their usage in different areas of modern society. Especially, in the industrial realm, the technical advancement in wireless network technology [1], Internet of Things (IoT) [2] and artificial intelligence have given rise to more progressive networked robotic applications that moves beyond their traditional deployment in production lines and deals with new challenges of industrial applications such as negotiation based decision-making, dynamic environmental disruptions, human–machine interaction and more personalized consumption demands etc. To combat all the complexity and diversity that it brings, significant hardware and software advancements are required for the robotic infrastructure. Unfortunately, there are still several limitation in the ways a robot hardware can actually be used or upgraded due to economical and functional constraints. As for the software advancements, this is where “cloud networked robotics” has made its mark. The introduction of cloud networked multi-robot system has merged the two ever-progressing domains of networked robotics and cloud computing. The added feature of cloud implies less dependence on human input and more support from ubiquitous resources [3] (i.e., computation, storage, crowdsourcing etc.).