Research Of Optimal Allocation Of Manufacturing Resource In T-Shaped Cloud Manufacturing Mode

Intelligent manufacturing is an essential procedure for realizing the national "Made in China 2025" strategy.It is also the main direction of promoting China’s manufacturing industry’s transformation to intelligentization and upgrading China from a quantity manufacturer to one of quality.The intelligent transformation of manufacturing enterprises is an important trend in developing the current manufacturing industry.However,relative to China’s huge enterprise base,the number that can truly achieve intelligent transformation is a tiny minority,especially the small and medium-sized enterprises(SMEs).As the lifeblood of our country’s economic development,SMEs contribute more than 50% of tax revenue and more than 60% of GDP.SMEs are the foundation for building a modern economic system and promoting high-quality economic development.However,SMEs are faced with problems such as single manufacturing resources,weak technical support,and simple manufacturing activities,which make such enterprises unable to carry out large-scale and complex manufacturing tasks.Therefore,the transformation of SMEs to intelligent manufacturing is constrained.With the development of network information technology,the cloud manufacturing mode is proposed to concentrate the manufacturing resources owned by various enterprises on the cloud platform.Through the optimal allocation of resources,enterprises can collaboratively complete complex manufacturing tasks.However,SMEs are limited by their own economic and technological level,so it is challenging to realize cloud manufacturing.Moreover,optimization allocation methods significantly impact the synergy effect of enterprises.Therefore,this paper conducts an in-depth study on the cloud manufacturing mode and resource optimization allocation method for SMEs.The main work and research results are as follows:(1)Proposed the T-shaped cloud manufacturing mode and the critical information model.For SMEs,based on the theory of Service Science and Cloud Manufacturing,the T-shaped cloud manufacturing mode is proposed and constructed with the guiding ideology of "horizontal management of decentralized resources,vertical coordination for horizontal services." In this mode,services are divided into horizontal service and vertical service from horizontal integration and vertical industry dimensions.The T-shaped cloud manufacturing mode promotes the collaboration of SMEs by combining vertical services to form high-quality horizontal services with the resources optimal allocation method.We define the essential factor of operation based on this mode,such as meta-resources,meta-services,etc.Moreover,we describe the operating mechanism,refine and establish the key information model under the T-shaped cloud manufacturing mode,provide process information for optimal allocation of resources.(2)Researched the mapping method of resource to service in T-shaped cloud manufacturing mode.For mapping resources to services in the T-shaped cloud manufacturing mode,namely,vertical service acquisition,the Division trend-means algorithm(Dt-means)is proposed.The Dt-means offers two essential concepts,the division trend degree(DTD)and the central radiation radius(CRR).The DTD is used to judge a subset whether the shape is quasi-circular and the distribution is uniform.The CRR is used to evaluate whether two different subsets belong to the same mergeable subset.The implementation process of the Dt-means is divided into three steps.Firstly,we use the k-means algorithm to divide the initial service set.Secondly,we use DTD to judge whether the service subsets need to be further divided until all the service sets are divided.Finally,we use CRR to evaluate whether two sets are the same mergeable subset.At last,through the experiments on the test dataset and the manufacturing resource dataset,it is verified that the Dt-means can realize the mapping of resources to services quickly,efficiently,and accurately.(3)Researched the evaluation method of service composition in T-shaped cloud manufacturing mode.The fuzzy evaluation model(FEM)is proposed based on the fuzzy representation of critical parameters,and the evaluation model based on parameter configuration(PCEM)is proposed.FEM,according to the characteristics of functional parameters representing the field advantages and the different fuzzy characteristics of the meta-task and meta-service parameters,evaluates the service composition from vertical service advantages.PCEM,according to the role of tasks’ parameter can guide the optimization direction of service composition,to evaluate service composition from horizontal service advantage by calculating the parameter sequence similarity between the service composition and the task.Through the experiments,the validities of the two proposed evaluation models are verified,which means the two evaluation models proved the decision basis for acquiring horizontal service from different T-shape aspects.(4)Researched the optimization method of services composition in T-shaped cloud manufacturing mode.Due to the problems of non-linear fitness distribution,neglected and decreasing excellent genes in the process of service composition optimization,in order to obtain the horizontal composition,we,based on the discrete fitness value distribution,multiple attributes of service composition,and the Pareto Principle characteristics of excellent genes of service composition,propose the Moving Window Flower Pollination Algorithm(MWFPA).MWFPA uses chained list to achieve linear traversal and multi-attribute optimization of the fitness distributions shape discrete compositions.In cross-pollination,the Moving Window is used to enhance the influence of excellent local genes,strengthen the overall optimization effect,and avoid falling into local extrema.Moreover,MWFPA continuously draws new genes into the population in self-pollination stage to improve the population’s excellent gene diversity.Through the simulation example,it is verified that the MWFPA can quickly and efficiently obtain better horizontal services.(5)Developed a cloud platform for optimal allocation of resources.We analyzed the current situation of small and medium-sized foundry enterprises,took the foundry industrial park as the representative,built a synergy platform.By building the model of the foundry industrial park,using the Java language-based Spring architecture and My SQL database,according to the structure of the T-shaped cloud manufacturing mode,the platform’s resource optimization allocation function is designed and implemented.The technologies studied in the previous chapters of this paper are applied to the platform.Finally,the platform functions are demonstrated.The feasibility of applying the T-shaped cloud manufacturing mode and resource optimization allocation in the foundry industry to achieve SMEs collaboration is verified through a casting example of an axle part.