Performance Evaluation of Enterprise Supply Chain

Research ArticlePerformance Evaluation of Enterprise Supply ChainManagementBased on the Discrete Hopfield Neural Network

Miaoling Bu

School of Business Shanxi Technology and Business College Taiyuan Shanxi 030032 China

Correspondence should be addressed to Miaoling Bu 20171072sxufeeducn

Received 29 June 2021 Revised 16 August 2021 Accepted 17 August 2021 Published 31 August 2021

Academic Editor Syed Hassan Ahmed

Copyright copy 2021Miaoling Buis is an open access article distributed under the Creative Commons Attribution License whichpermits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

In order to make up for the shortcomings of current performance evaluation methods this paper proposes a new method ofenterprise performance evaluation discusses the construction principle of the evaluation index and proposes a method ofenterprise supply chain overall performance evaluation based on the discrete Hopfield neural network (DHNN) algorithmEnterprise supply chain (SC) is an important way for enterprises to conduct business with other strategic partners in themarket and the improvement of SC performance is an important way to improve the core competitiveness of enterprisesso it is of great value to study the performance evaluation and index design of the enterprise SC is method calculates thelevel value of the overall performance of the SC is level value is a value between 0 and 1 e higher the value the higherthe overall performance level of the SC erefore when evaluating the overall performance of the SC appropriate indexweights must be selected according to the characteristics of the industry which helps to objectively evaluate the overallperformance of the SC

1 Introduction

With the rapid development of science and technology thecontinuous deepening of global economic integration hasgreatly increased the intensity of market competition Inorder to keep up with the pace of social development en-terprises continue to shift their sights to the supply chainmanagement (SCM) model with ldquohorizontal integrationrdquo asthe core concept [1] Under the influence of various factorssuch as consumer demand diversification and product di-versity SCM has been unanimously praised by manycompanies due to its many advantages and the SC com-posed of multiple companies has been continuously de-veloped [2] In todayrsquos economic globalization and flexiblemarket demand environment node companies in the SC canbe scattered all over the world e business activities ofenterprises are decentralized and the performance evalua-tion of enterprises within the SC unifies the decentralizedoperations under the same standard [3] Enterprise per-formance evaluation refers to the use of quantitative sta-tistics and operation research methods the use of a specific

indicator system and a unified evaluation standard in ac-cordance with certain procedures through quantitative andqualitative analyses to evaluate the operating benefits andoperators of the enterprise during a certain operating periodIt is science to make an objective fair and standard com-prehensive evaluation to truly reflect the actual situation ofthe enterprise and predict the future development prospectsof the enterprise e content of the evaluation mainly in-cludes the profitability of the company the level of assetoperation the ability to repay debts and the ability to followup development [4]

Supplier selection is an important part of SCM Atpresent the research on supplier selection is more and moreextensive and in depth [5] Selecting the best supplier canreduce the overall system cost improve competitivenessand enable enterprises to find partners to solve problems anddevelop together Many large international enterprises havejoined suppliers in the research and development of newproducts seeking to achieve ldquowin-winrdquo effect in the com-petition of multiple SCs [6] Previous studies mainly focusedon the analysis and selection of various attributes of

HindawiComputational Intelligence and NeuroscienceVolume 2021 Article ID 3250700 11 pageshttpsdoiorg10115520213250700

suppliers while the research on supplier evaluation methodsand evaluation models only gradually increased in recentyears [7] e evaluation method of Siying using the BPnetwork is proposed by some people Data envelopmentanalysis (DEA) is applied to supplier evaluation [8] Somescholars use the analytic hierarchy process in the supplierevaluation process [9] Researchers apply activity-based costanalysis (ABC) to evaluate suppliers [10]erefore it is veryimportant to develop a simple effective and practicalsupplier evaluation model Combining the traditional ana-lytic hierarchy process this paper proposes and establishesan enterprise SCM performance index evaluation modelbased on the discrete Hopfield neural network (DHNN)Using the learning objective and dynamic characteristics ofthe DHNN we can solve the problem of enterprise supplychain performance index evaluation

DHNN is a single-layer feedback network composed oftwo neurons interconnected e input and output ofneurons are discrete values of 1 or minus1 which represent theactivation and inhibition states of neurons respectively [11]Stability evolution ability storage capacity domain of at-traction and convergence rate are the five indicators tomeasure the performance of this neural network [12] For along time cost effectiveness determines the success of anenterprise to a large extent e main way to evaluate theperformance of an enterprise is to compare its profitabilityand market share with competitors through financial in-dicators [13] Comparing with the performance of theleading SC in the same industry can objectively reflect theperformance level of the SC and the competitive position ofthe SC in the market [14] erefore the benchmark eval-uation method of the overall performance of the SC issuitable for the SC which is in the position of a challengerand follower in the industrye brand-new enterprise SCMtheory requires all enterprises to redesign the performanceevaluation system and further explore the sustainable de-velopment ability of enterprises to keep up with the times[15] erefore the scientific and effective evaluation of theenterprise performance of the SCM theory has turned intoan important topic for enterprises and researchers It can beseen that the research on the enterprise performance eval-uation of SCM theory has very important practicalsignificance

2 Related Work

Performance management is an effective tool to achievestrategic goals Performance management indicators shouldbe decomposed around strategic objectives rather thanrelated to the implementation of strategic objectives Onlywhen the employeesrsquo efforts are consistent with the com-panyrsquos strategic goals can the companyrsquos overall performancebe improved Kazancoglu et al [16] proposed that the designof the evaluation index system should follow several basicprinciples Darvish et al [17] pointed out when discussingldquohow to construct a successful performance evaluation indexsystemrdquo that the goal of performance evaluation is not toobjectively quantify all aspects but to avoid subjective as-sumptions doubts and measurements in the evaluation

process Raut et al [18] pointed out that the competitionbetween enterprises is gradually transforming into thecompetition between the SC and the SC Companies withforward-looking concepts are reviewing a series of processesfrom raw materials manufacturing transportation anddistribution to customers erefore a feasible corporateperformance evaluation should be a realistic evaluation ofthe pros and cons of the overall SC management Peng et al[19] pointed out that in the process of compiling perfor-mance evaluation indicators it is necessary to communicateand decompose the strategic objectives of the enterprise instages endow each position of the enterprise with strategicresponsibilities and ensure that each employee performs hisor her duties He [20] pointed out that some small- andmedium-sized enterprises have a small scale of their owndevelopment lack of investment in technological develop-ment and insufficient innovation capabilities which in turncaused the slow development of enterprise technology de-velopment and innovation capabilities and it was difficult toachievemanagement innovation Duan [21] showed that dueto the limited strength of some companies under thepremise of uncoordinated information they usually chooseto compromise with the conditions of other companies inthe SC or are constrained by phenomenon conditions In theprocess of selecting cooperative companies due to their lackof strength and scale they usually choose companies withrelatively low prices or other companies with unsatisfactoryconditions

Wu et al [22] proposed a method to automaticallyextract text features based on the convolutional neuralnetwork Njitacke et al [23] applied the linear aug-mentation method in controlling the multistable HNNsinto a monostable network in [24] in order to improvethe structure of the RHNN model the collective guidancefactor-based pathfinder algorithm has been proposedWu et al [25] collected vast online oil news and used theconvolutional neural network to extract relevant infor-mation automatically for forecasting US oil marketsSatisfactory performance was obtained Hu and Qin [26]pointed out that the degree of cooperation betweensuppliers in improving quality can be measured by thepercentage of suppliersrsquo effective participation in thequality improvement activities of buyers and manufac-turers to the total participation e larger the index thehigher the cooperation degree between suppliers andbuyers and manufacturers and the stronger the supplycoordination ability Chen [27] showed that because ofthe complexity of SCM it is difficult to evaluate with asingle index and it is necessary to evaluate from multipleangles and perspectives and establish a hierarchical indexsystem According to different specific problems anddifferent evaluation contents and objectives the estab-lished index system is also different Zhu and Luo [28]showed that each enterprise SC can be used as a sub-system in the current market environment and the op-eration mode of different SCs will be different Howeverevery SC will transmit the market demand and infor-mation of the external environment to the inside of theSC and then each SC will respond to the external demand

2 Computational Intelligence and Neuroscience

and market information according to its own situationand change the existing and external environmentaccording to the actual situation Contradictory orga-nizational structure or evaluation standards make it morereasonable and easier to operate so that the SC canoperate relatively smoothly and continuously in thecurrent market environment changes

3 Performance Evaluation Index of theEnterprise SC

31 Overview of Performance Management Managementperformance refers to the management process that en-sures that employeesrsquo work activities and performance areconsistent with the organizationrsquos goals and throughcontinuous communication to make further contribu-tions to the realization of the organizationrsquos goals Per-formance management is a three-to-one system consistingof preplanning in-process management and posteventevaluation Performance appraisal refers to the evaluationof a subjectrsquos work objectives or performance standardsand the use of scientific evaluation methods for evalua-tion e feedback process of employees on their workcompletion and performance affects their evaluation re-sults Performance management and performance eval-uation are obviously different Performance evaluation isonly a link in the process of performance managementand performance evaluation cannot replace performancemanagement [29 30] Performance management is thecore of corporate management activities as well as thecore of human resource management and developmentPerformance management indicators are the process ofjudging the work performance and creating value ofmanagement personnel at all levels who implement thebusiness processes and results of the enterprise throughthe unit or method of clear performance appraisal goalsFor example company employees implement companyemployee evaluation indicators including comprehensiveinspection and evaluation of company employeesrsquo ethicswork performance abilities and attitudes to determinework performance and potential management practicesEvaluate the companyrsquos business policy managementphilosophy existing organizational structure and allsystems related to the companyrsquos performance manage-ment Design the details of the system cycle including therationality of the original performance appraisal systemthe current employee compensation and benefits planreward and punishment system and labor managementregulations Figure 1 shows the cyclical process of per-formance management

With the intensification of market competition enter-prises are increasingly demanding higher-level manage-ment technical and other types of talents At this timewestern modern performance management concepts sys-tems and tools began to be gradually introduced into ChinaAmong them key performance indicators and balancedscorecard as advanced strategy-oriented performancemanagement methods are being paid attention to andapplied by more and more enterprises

32 SCM Overview SCM is an advanced enterprise man-agement concept which promotes the full integration of newideas and technologies in modern management As an ad-vanced management mode SCM is characterized by codingagility and integration At the present stage from the un-derstanding of the meaning of SCM from a relatively unifiedperspective SCM refers to enterprisesrsquo understanding of thedevelopment laws and interrelationships of different links inthe SC With the help of management planning organiza-tion coordination and encouragement a whole set of linkswill be integrated into the production and operation ofproducts to maximize the interests of enterprises e basicidea of SCM is to promote the integration of core com-petitiveness of enterprises Especially through internal SCMit can promote the integration of key competitive sub-projects within enterprises and promote the development ofthe SC rough external SCM it can integrate the corecompetitiveness of terminal enterprises In other words fordifferent enterprises in the SC total SCM mainly includesinternal SCM and external SCM to determine the effec-tiveness of existing systems or compare different choicesAfter the performance evaluation index is determined theappropriate evaluation method should be selected accordingto the evaluation purposee differences between SCM andtraditional management mode are mainly manifested inseveral aspects SCM has essential characteristics SCM hasmacronature focusing on strategic management SCMpromotes the optimization and integration of resourcesamong different enterprises in the SC and promotes theeffective utilization of resources of a group of enterprisesSCM has higher pursuit than the traditional managementmode

33 Construction of theOverall Performance Evaluation IndexSystem of the Enterprise SC e key to evaluate the overallperformance of the SC scientifically and objectively is theperformance appraisal index system When evaluating theoverall performance of the SC it is important to consider thefinancial performance and response time of the SC eflexibility of SC operation should be considered from adynamic point of view which is the content of businessprocess improvement and innovation among SC memberenterprises erefore the following requirements are putforward to evaluate the overall performance of the SCcomprehensive management of the overall performance ofthe SC and analyzing performance problems from theperspective of the whole SC not just analyzing performanceproblems Look at the SC from the perspective of a specificcompanye overall performance of the SC should focus onthe future development of the organization and strengthenthe feedforward of performance management Besidesevaluating the performance of core links we should alsoconsider evaluating the performance of noncore linksNonfinancial indicators and financial indicators are equallyconcerned Scientific index system is the basis of objectivelyand accurately evaluating the overall performance of the SCFrom the perspective of SC function management a mul-tilevel performance evaluation index system of the SC is

Computational Intelligence and Neuroscience 3

established including organizational management It con-sists of 6 first-level indicators and 23 second-level indicatorsincluding information management demand managementlogistics management operation management and mar-keting management e index system for evaluating theoverall performance of the SC is shown in Table 1

With reference to Kaplan and Nortonrsquos balancedscorecard theory the following enterprise performanceevaluation index system based on SCM theory is designed asshown in Figure 2

4 Performance Evaluation Index andCoordination Measures of CompanySC Planning

41 SC Planning Performance Evaluation Indicators As aresult of the SC planning stage companies must clearlyformulate a set of operational strategies that govern theshort-term operation of the SCe SC structure determinedin the strategic phase determines the work to be completedin the SC planning phase In the planning stage wemust firstpredict the demand in different markets in the coming yearthen determine the market sources and inventory levels signmanufacturing subcontracts and determine goods replen-ishment and inventory policies emergency strategies forshortages and the timing and scope of marketing

411 Product Development Product development efficiencycan be measured by the product development cycle Productdevelopment cycle refers to the time from the emergence ofmarket opportunities to the production of new products inthe SC and the acquisition of sales revenue Specifically itincludes t1 from the emergence of market opportunities tothe discovery of market opportunities t2 from the discoveryof market opportunities to the success of research and de-velopment t3 from the success of research and development

to the launch of new products and t4 from the launch of newproducts to the acquisition of sales revenue erefore thecalculation formula of the product development cycle andTpd is as follows

Tpd 11139444

i1ti (1)

412 Supply e degree of cooperation between suppliersin improving quality can be measured by the percentage ofsuppliersrsquo effective participation in the quality improvementactivities of buyers and manufacturers to the total partici-pation e larger the index the higher the cooperationdegree between suppliers and buyers and manufacturers andthe stronger the supply coordination abilitye data neededfor index calculation can be obtained from the suppliermanagement database Assuming thatm suppliers supply forbuyers (manufacturers) the number of times that supplierF

jig effectively participates in the quality improvement ac-

tivities of buyers in time T is Fjtig and the total number of

participants is Fjtig the calculation formula of the joint

cooperation degree Rmc of quality improvement is

Rmc 1113936

mj1 F

jiq

1113936mj1 F

jtiq

times 100 (2)

413 Delivery e indicators of emergency distributionresponse degree and distribution reliability can be used todescribe the performance of distribution coordination indetail

Emergency delivery response level it reflects the re-sponse capacity of delivery which can be measured by thepercentage of the number of emergency deliveries achievedand the number of emergency deliveries required Assuming

Organizational goals decompose workunit functions

Performance indicator system

Performance plan

Performance feedback Performance implementationand management

Performance Evaluation

Performance results

Figure 1 e cyclical process of performance management


that the number of emergency deliveries realized by thedistributor during the period T is Ffud and the requirednumber of emergency deliveries is Fud the formula forcalculating Rrud of the emergency delivery response degreeof the distributor during this period is

Rrud Ffud

Fudtimes 100 (3)

Distribution reliability it reflects the ability of effectivedistribution which can be measured by the percentage of thenumber of times of distribution realized according to thespecified requirements and the number of times of planneddistribution Suppose that the number of times of distri-bution achieved by the distributor according to the specifiedrequirements in the period T is Ffqd and the number of timesof planned distribution in the period T is Fqd then thecalculation formula of distribution reliability Fdd of thedistributor in the period T is as follows

Fdd Ffqd

Fq d

times 100 (4)

In the self-made program the initial value is set to zerothe number of network nodes is changed and the averageerror of the network is compared e obtained data areshown in Table 2 It can be seen from Figure 3 that when thenumber of hidden layer nodes is 6 theMSE value reaches theminimum of 747Eminus 06

Figure 4 shows the changes in the inventory value of acertain month in the next week e result is compared withthe safety stock and the maximum stock It can be concludedthat the stock on Tuesday is out of specification and thestock is short on Saturday and the corresponding eventwarning is generated

rough analysis there is a linear relationship betweenthe number of SC outlets and logistics cost and labor coste correlation analysis of relevant data from 2016 to 2020verifies the existence of this rule as shown in Figures 5 and 6

e logistics efficiency of enterprises with the back-ground of the transportation industry is an important aspectof enterprise productivity e logistics cost and manpower

cost reduced by the scale benefit of the logistics platformwhich showed a great performance of enterprise produc-tivity improvement

42 Coordination Measures In the increasingly compet-itive global economic environment the evaluation ofenterprise SC innovation and learning ability is be-coming more and more important SC innovation andlearning ability is one of the concrete manifestations ofthe core competitiveness of an enterprise and it is alsothe fundamental guarantee for the long-term prosperityand progress of the enterprise e companyrsquos salesdepartment production workshop and other core de-partments are more and more likely to ignore andcomplain about other supporting or service departmentsAnd these phenomena restrict the maximization of en-terprise efficiency from another aspect Many enter-prisesrsquo internal gold mines violate the mining and wastehuge resources Especially for the enterprises establishedfor more than 10 years the core production equipmenttends to be aging and often breaks down resulting inshort-term shutdown production parts cannot be sup-plied in time subsequent processes cannot be carried outsmoothly or multiple orders are executed at the sametime in the peak sales season and the existing productioncapacity of the production workshop cannot fully copewith it or there are business or technical adjustments inthe execution of a project e mutation reaction cannotbe adjusted in time resulting in confusion in the pro-duction process unable to operate normally When newproblems arise it is necessary to further adjust the or-ganizational structure of the companyrsquos SC establish anintegrated management department and improve andperfect the coordination mechanism of the division ofresponsibilities among departments which can bettersolve the problem of business assistance among de-partments in the SC prevent mutual prevarication andinaction caused by overlapping responsibilities or poorconnection between front and back and further improvework efficiency and promote the healthy development of

Table 1 SC performance evaluation index system

e overall performance evaluation index system of the SCFirst-levelindicators

Organization and managementindicators Information management indicators Demand management indicators

Secondaryindicators

1 Network formation cycle 1 e degree of integratedmanagement 1 Demand forecast accuracy

2 Network reorganization capability 2 Level of information sharing 2 Customer satisfaction3 Internal transaction fee rate 3 Information feedback cycle 3 Demand response cycle

4 SC coordination cost 4 Customer relationshipmanagement

First-levelindicators Logistics management indicators Operation management indicators Marketing management indicators

Secondaryindicators

1 e degree of logistics integration 1 Unit product cost 1 Unit product marketing cost2 On-time supply rate 2 Planned order completion rate 2 Market share3 100 yuan marketing expense rate 3 Production cycle 3 Per capita sales4 Out-of-stock rate 4 Capacity utilization

5 Labor productivity


Enterprisedevelopment vision

and strategy

Supply chain business process

The relationship betweenthe upper and lower nodes

of the supply chain

Supply chain innovation andlearning ability

Production-sales rate

Production demand rate

Product production or servicecycle

Total operating cost

Cost profit margin

Product quality qualificationrate

On-time delivery rate

After-sales service quality

Profit growth rate

Human capital ratio

New service revenue ratio

Employee recommendedgrowth rate

Growth rate of total employeetraining hours

Investigateeach upperand lower

node

Sales profit margin

Comparable product costreduction rate

Inventory turnover

Growth rate of total outputvalue

Accounts Receivable TurnoverRate

Supply chain economics

Figure 2 Enterprise performance evaluation index system based on SCM theory


various business works in the SC Coordinate the rela-tionship between departments and other departments asshown in Figure 7

5 Evaluation of the Performance Index ofEnterprise SCM Based on the DHNN

51 DHNN Hopfield neural network (HNN) is a single-layer binary neuron recurrent network which has dynamicfeedback from the system output to the input process Itstopology is shown in Figure 8

HNN is a kind of fully connected and feedback networkincluding discrete and continuous e evolution of theHNN state is a complex nonlinear dynamic system estability of the system can be analyzed by ldquoenergy functionrdquoUnder certain conditions the energy of the network isconstantly reduced and finally converges to the stabilitypoint of the system If the stable point of the system isregarded as a memory the process of flowing from the initialstate to the stable point is the process of searching for thememory DHNN is mainly used for associative memoryWhen the input vector I is taken as an initial value thenetwork evolves through feedback and obtains a vector Vfrom the output of the network V is a stable memory related

Table 2 e influence of the number of hidden layer nodes on thenetwork

Layer number 2 4 6 8 10 12 14Epoch 177 99 54 42 45 42 41MSE (10minus 6) 974 767 747 985 785 828 931Layer number 2 4 6 8 10 12 14

2 4 6 8 10 12 14

120E-05

100E-05

800E-06

600E-06

400E-06

200E-06

000E+00

200

180

160

140

120

100

80

60

40

20

0

Epoc

h

MSE

MSEEpoch

Figure 3 Schematic diagram of the influence of the number ofhidden layer nodes on the network

Mon Tues Wed Thur Fri Sat Sun

140

120

100

80

60

40

20

0

Time

Inve

ntor

Maximum inventoryDaily inventorySafety stock

Figure 4 Product inventory changes

Correlation analysis of logistics cost and thenumber of business centers

Business center2502252001751501251007550250

Logi

stics

cost

()

600

500

400

300

200

100

000

Linear (logistics cost)Logistics cost

Figure 5 Correlation analysis of logistics cost and the number ofbusiness centers

Art

ifici

al co

st (

)

Business center

600

500

400

300

200

100

0002502252001751501251007550250

Artificial costLinear (logistics cost)

Figure 6 Correlation analysis of labor costs and the number ofbusiness centers


to the evolution of the initial value i In order to use it tosolve the problem of provider evaluation we first need todesign W and I so that the samples of memory patternscorrespond to the stable points in the network is isequivalent to training a neural network process etrained memory model corresponds to the standard levelevaluated by the provider and the data to be evaluated aretaken as the new initial state e network regards theinitial state as a new prompt mode and recalls the ldquolatestrdquostate is is the process of associative memory

e process of evaluation with the Hopfield network isdivided into memory and association processes e outerproduct design can be divided into the following steps

(1) According to the sample V1 VM that needs to bememorized the weight is calculated as follows

wij 1113944

m

k1V

ki V

kj ine j

0 i j

⎧⎪⎪⎨

⎪⎪⎩(5)

Ij 0 Ij is the initial input of the network(2) Let the test sample be the initial value of the network

output and let the vector be an arbitrary input vectorso that

V t0( 1113857 Vl V isin R

n (6)

(3) Use the following iterative formula to calculate

Vi(t + 1) sgn 1113944n

j1wijVj(t)⎛⎝ ⎞⎠ (7)

Plan

Purchase

Technology

Produce

SalesCoordination

Figure 7 e relationship between the coordination department and other departments

w11

w12

w13

w31

w32

w33

w21 w23w22

x1

y1 y2 y3

x2 x3

Layer 0

Layer 1

Figure 8 Discrete Hopfield network structure


(4) Repeat the iteration until each cell remainsunchanged

Vi(t + 1) Vj(t) (8)

At this time Vl returns to a memory sample that hasbeen learned Because each evaluation index has differentweights the influence of weights must be considered whenevaluating

For the optimal control problem of the bilinear discrete

system take A

0 15 051 0 12 0 08

⎡⎢⎢⎢⎢⎢⎣⎤⎥⎥⎥⎥⎥⎦ B

011

⎡⎢⎢⎢⎢⎢⎣⎤⎥⎥⎥⎥⎥⎦ Q

1 0 00 1 00 0 1

⎡⎢⎢⎢⎢⎢⎣⎤⎥⎥⎥⎥⎥⎦

R D 1x(0) 08 05 031113858 1113859T and ϕ 0 and the sim-

ulation diagram obtained by HNN optimization is shown inFigures 9 and 10

Figures 9 and 10 show the curves of the optimal state andoptimal control respectively It can be seen from the graphthat the system state converges through 13 steps of iteration

52 Overall Performance EvaluationMethod of the EnterpriseSC e overall performance evaluation of the enterprise SCis a classification problem so it is necessary to establish anevaluation index system collect relevant data and use thesimulation system to analyze and process the data in order toobtain objective and accurate evaluation results ere aretwomainmethods to evaluate the overall performance of theSC self-evaluation method and benchmarking method eself-evaluation method of the overall SC performancecompares the overall actual performance of the SC with theoverall target performance of the SC to evaluate the overallperformance level of the SCis method calculates the levelvalue of the overall performance of the SCis level value isa value between 0 and 1 e higher the value the higher the

overall performance level of the SC In order to directlyreflect the accuracy of its fitting refer to Figure 11

In order to make the curve simulation more accurate ahigher number of times can be adopted as shown inFigure 12 which is a sixth-order simulation curve withvery good accuracy but it is very difficult to solve it inoptimization calculation It is divided into the arithmeticaverage method and weighted average method e self-evaluation method is applicable to the SC of the leader inthe industry e benchmarking method of the SC overallperformance is to compare the actual performance of thewhole SC with the performance of the leading SC in thesame industry to evaluate the relative performance level ofthe whole SC Due to the subjective randomness of theoverall performance self-evaluation of the SC the relativeevaluation of the overall performance of the SC is moremeaningful

y2520151050

x

08

06

04

02

0

-02

-04

X1X2X3

Figure 9 Optimal state x

0 2 4 6 8 10 12 14 16 18 20

05

0

-05

-1

y

u

Figure 10 Optimal control u

Time index

Fitti

ng ac

cura

cy in

dex

360

330

300

270310280250220190160

Enterprise 3Enterprise 2Enterprise 1

Figure 11 Quadratic simulation curve


6 Conclusions

With the continuous development of social economy thetraditional evaluation methods can hardly meet the re-quirements of enterprise performance evaluation under thecurrent economic environment so it becomes more andmore critical to build enterprise performance evaluationbased on SCM theory Performance management is not onlythe core work of human resource management but also oneof the core competitiveness of the company in an invincibleposition in the fierce market competition DHNN usespattern association the training process is simple and doesnot require a large number of samples and the networkresponse time is very short the pattern adopted by DHNNusually can only complete 1 to 2 iterations especially suitablefor supplier evaluation and grading Research on the overallperformance evaluation of the SC is the basis for improvingthe performance of the SC is article establishes theDHNN-based enterprise SCM performance index evalua-tion e selection of performance indicators and weightswill inevitably be different due to different industries of theSC itself erefore when evaluating the overall perfor-mance of the SC we must select the appropriate indexweights in combination with the characteristics of the in-dustry so as to objectively evaluate the overall performanceof the SC Although the associative memory function of theneural network is very strong it also has some defectsBecause the associative memory ability is restricted by thememory capacity and sample differences when there aremany memory patterns which are easy to be confused thenetwork cannot distinguish the correct pattern well and thestable state is often not the memory pattern Moreover allmemory patterns are not recalled with the same memoryintensity We can foresee that with the further improvementand development of artificial intelligence that is neuralnetwork system its advantages and achievements will not

only be reflected in the overall optimization calculation butalso be developed and expanded due to its uniqueadvantages

Data Availability

All the data in this paper come from the data statistics of thetest process All the data are real and can be used

Conflicts of Interest

e author declares that there are no conflicts of interest

Acknowledgments

is work was supported by Shanxi Technology and Busi-ness College

References

[1] V K Sharma P Chandna and A Bhardwaj ldquoGreen SCMrelated performance indicators in agro industry a reviewrdquoJournal of Cleaner Production vol 141 pp 1194ndash1208 2017

[2] A Garai and T K Roy ldquoMulti-objective optimization of cost-effective and customer-centric closed-loop supply chainmanagement model in T-environmentrdquo Soft Computingvol 24 no 1 pp 155ndash178 2020

[3] A P Tafti J Badger E LaRose et al ldquoAdverse drug eventdiscovery using biomedical literature a big data neural net-work adventurerdquo JMIR Medical Informatics vol 5 no 4p e51 2017

[4] Y Yu and C H Fu ldquoResearch on associative memory abilitybased on discrete hopfield neural networkrdquo Software Guidevol 15 no 9 pp 146ndash148 2016

[5] L P Zhu F H Wang and H Q Li ldquoDiscrete hopfield neuralnetwork learning algorithmrdquo Computer Engineering andDesign vol 39 no 3 pp 831ndash835 2018

[6] J C Guo ldquoEvaluation of the overall performance of theenterprise SC based on hopfield neural networkrdquo LogisticsEngineering and Management vol 41 no 7 pp 140ndash1432019

[7] D Li and A Nagurney ldquoSC performance assessment andsupplier and component importance identification in ageneral competitive multitiered SC network modelrdquo Journalof Global Optimization vol 67 no 1-2 pp 223ndash250 2017

[8] H B Singhry A A Rahman and N S Imm ldquoSC innovationand performance of manufacturing companiesrdquo InternationalJournal of Advanced Manufacturing Technology vol 86 no 1-4 pp 663ndash669 2016

[9] G Lu X Koufteros and L Lucianetti ldquoSC security a clas-sification of practices and an empirical study of differentialeffects and complementarityrdquo IEEE Transactions on Engi-neering Management vol 64 no 2 pp 234ndash248 2017

[10] W Mu C E Van Middelaar J M Bloemhof B Engel andI J M de Boer ldquoBenchmarking the environmental performanceof specialized milk production systems selection of a set of in-dicatorsrdquo Ecological Indicators vol 72 pp 91ndash98 2017

[11] G L Tortorella R Miorando and G Marodin ldquoLean SCMempirical research on practices contexts and performancerdquoInternational Journal of Production Economics vol 193pp 98ndash112 2017

[12] M-C Yu Y-C J Wu W Alhalabi H-Y Kao andW-H Wu ldquoResearchGate an effective altmetric indicator for

Time index310280250220190160

Fitti

ng ac

cura

cy in

dex

360

350

340

330

320

310

300

390

280


Figure 12 Sixth simulation curve


active researchersrdquo Computers in Human Behavior vol 55pp 1001ndash1006 2016

[13] Y Xu and X J Meng ldquoRethinking domestic enterprise SCMrdquoMarketWeekly No6eoretical Research vol 10 pp 46-47 2016

[14] Y Y Ouyang and R W Li ldquoEnterprise performance eval-uation model based on SC integrationrdquo ModernManufacturing Engineering vol 462 no 3 pp 50ndash57 2019

[15] J Liu Y Guo and J Y Gu ldquoResearch on the performanceevaluation index system of green SCM system in printingindustryrdquo Information Technology and Standardizationvol 415 no 7 pp 69ndash73 2019

[16] Y Kazancoglu I Kazancoglu and M Sagnak ldquoA new holisticconceptual framework for green SCM performance assess-ment based on circular economyrdquo Journal of Cleaner Pro-duction vol 195 pp 1282ndash1299 2018

[17] M Darvish C Archetti and L C Coelho ldquoTrade-offs be-tween environmental and economic performance in pro-duction and inventory-routing problemsrdquo InternationalJournal of Production Economics vol 217 pp 269ndash280 2019

[18] R D Raut S Luthra B E Narkhede et al ldquoExamining theperformance oriented indicators for implementing greenmanagement practices in the Indian agro sectorrdquo Journal ofCleaner Production vol 215 pp 926ndash943 2019

[19] L Peng M Peng B Liao et al ldquoe advances and challengesof deep learning application in biological big data processingrdquoCurrent Bioinformatics vol 13 no 4 pp 352ndash359 2018

[20] D H He ldquoResearch on enterprise performance evaluationbased on SCM theoryrdquo Financial Economics vol 4 no 478pp 95ndash97 2018

[21] L J Duan ldquoResearch status of SC performance evaluation inthe new erardquo Market Modernization vol 14 pp 48-49 2016

[22] B Wu L Wang S X Lv et al ldquoEffective crude oil priceforecasting using new text-based and big-data-driven modelrdquoMeasurement vol 168 Article ID 108468 2021

[23] Z T Njitacke S D Isaac T Nestor et al ldquoWindow ofmultistability and its control in a simple 3D hopfield neuralnetwork application to biomedical image encryptionrdquoNeuralComputing and Applications vol 33 no 10 pp 1ndash20 2021

[24] M Yang L Zhang T-Y Li N Yousefi and Y-K LildquoOptimal model identification of the PEMFCs using opti-mized rotor hopfield neural networkrdquo Energy Reports vol 7no 1 pp 3655ndash3663 2021

[25] B Wu L Wang S Wang et al ldquoForecasting the US oil marketsbased on social media information during the COVID-19pandemicrdquo Energy vol 226 Article ID 120403 2021

[26] Z Hu and W Qin ldquoFuzzy method and neural network modelparallel implementation of multi-layer neural network basedon cloud computing for real time data transmission in largeoffshore platformrdquo Polish Maritime Research vol 24 no s2pp 39ndash44 2017

[27] D Chen ldquoResearch on traffic flow prediction in the big dataenvironment based on the improved RBF neural networkrdquoIEEE Transactions on Industrial Informatics vol 13 no 4pp 2000ndash2008 2017

[28] Y Z Zhu and M L Luo ldquoResearch on enterprise logisticsperformance evaluation index systemrdquo Business Managervol 540 no 8 pp 72-73 2018

[29] Y H Zhang and J Q Peng ldquoConstruction of the performanceevaluation index system for SC collaborative innovationrdquoSocial Scientist vol 10 pp 71ndash75 2016

[30] S A Saeed Alzaeemi S Sathasivam and M Velavan ldquoAgent-based modeling in doing logic programming in fuzzy hopfieldneural networkrdquo International Journal of Modern Educationand Computer Science vol 13 no 2 pp 23ndash32 2021


suppliers while the research on supplier evaluation methodsand evaluation models only gradually increased in recentyears [7] e evaluation method of Siying using the BPnetwork is proposed by some people Data envelopmentanalysis (DEA) is applied to supplier evaluation [8] Somescholars use the analytic hierarchy process in the supplierevaluation process [9] Researchers apply activity-based costanalysis (ABC) to evaluate suppliers [10]erefore it is veryimportant to develop a simple effective and practicalsupplier evaluation model Combining the traditional ana-lytic hierarchy process this paper proposes and establishesan enterprise SCM performance index evaluation modelbased on the discrete Hopfield neural network (DHNN)Using the learning objective and dynamic characteristics ofthe DHNN we can solve the problem of enterprise supplychain performance index evaluation

DHNN is a single-layer feedback network composed oftwo neurons interconnected e input and output ofneurons are discrete values of 1 or minus1 which represent theactivation and inhibition states of neurons respectively [11]Stability evolution ability storage capacity domain of at-traction and convergence rate are the five indicators tomeasure the performance of this neural network [12] For along time cost effectiveness determines the success of anenterprise to a large extent e main way to evaluate theperformance of an enterprise is to compare its profitabilityand market share with competitors through financial in-dicators [13] Comparing with the performance of theleading SC in the same industry can objectively reflect theperformance level of the SC and the competitive position ofthe SC in the market [14] erefore the benchmark eval-uation method of the overall performance of the SC issuitable for the SC which is in the position of a challengerand follower in the industrye brand-new enterprise SCMtheory requires all enterprises to redesign the performanceevaluation system and further explore the sustainable de-velopment ability of enterprises to keep up with the times[15] erefore the scientific and effective evaluation of theenterprise performance of the SCM theory has turned intoan important topic for enterprises and researchers It can beseen that the research on the enterprise performance eval-uation of SCM theory has very important practicalsignificance

2 Related Work

Performance management is an effective tool to achievestrategic goals Performance management indicators shouldbe decomposed around strategic objectives rather thanrelated to the implementation of strategic objectives Onlywhen the employeesrsquo efforts are consistent with the com-panyrsquos strategic goals can the companyrsquos overall performancebe improved Kazancoglu et al [16] proposed that the designof the evaluation index system should follow several basicprinciples Darvish et al [17] pointed out when discussingldquohow to construct a successful performance evaluation indexsystemrdquo that the goal of performance evaluation is not toobjectively quantify all aspects but to avoid subjective as-sumptions doubts and measurements in the evaluation

process Raut et al [18] pointed out that the competitionbetween enterprises is gradually transforming into thecompetition between the SC and the SC Companies withforward-looking concepts are reviewing a series of processesfrom raw materials manufacturing transportation anddistribution to customers erefore a feasible corporateperformance evaluation should be a realistic evaluation ofthe pros and cons of the overall SC management Peng et al[19] pointed out that in the process of compiling perfor-mance evaluation indicators it is necessary to communicateand decompose the strategic objectives of the enterprise instages endow each position of the enterprise with strategicresponsibilities and ensure that each employee performs hisor her duties He [20] pointed out that some small- andmedium-sized enterprises have a small scale of their owndevelopment lack of investment in technological develop-ment and insufficient innovation capabilities which in turncaused the slow development of enterprise technology de-velopment and innovation capabilities and it was difficult toachievemanagement innovation Duan [21] showed that dueto the limited strength of some companies under thepremise of uncoordinated information they usually chooseto compromise with the conditions of other companies inthe SC or are constrained by phenomenon conditions In theprocess of selecting cooperative companies due to their lackof strength and scale they usually choose companies withrelatively low prices or other companies with unsatisfactoryconditions

Wu et al [22] proposed a method to automaticallyextract text features based on the convolutional neuralnetwork Njitacke et al [23] applied the linear aug-mentation method in controlling the multistable HNNsinto a monostable network in [24] in order to improvethe structure of the RHNN model the collective guidancefactor-based pathfinder algorithm has been proposedWu et al [25] collected vast online oil news and used theconvolutional neural network to extract relevant infor-mation automatically for forecasting US oil marketsSatisfactory performance was obtained Hu and Qin [26]pointed out that the degree of cooperation betweensuppliers in improving quality can be measured by thepercentage of suppliersrsquo effective participation in thequality improvement activities of buyers and manufac-turers to the total participation e larger the index thehigher the cooperation degree between suppliers andbuyers and manufacturers and the stronger the supplycoordination ability Chen [27] showed that because ofthe complexity of SCM it is difficult to evaluate with asingle index and it is necessary to evaluate from multipleangles and perspectives and establish a hierarchical indexsystem According to different specific problems anddifferent evaluation contents and objectives the estab-lished index system is also different Zhu and Luo [28]showed that each enterprise SC can be used as a sub-system in the current market environment and the op-eration mode of different SCs will be different Howeverevery SC will transmit the market demand and infor-mation of the external environment to the inside of theSC and then each SC will respond to the external demand















Tpd 11139444

i1ti (1)






Rmc 1113936

mj1 F

jiq

1113936mj1 F

jtiq

times 100 (2)





Performance plan



Performance results




Rrud Ffud

Fudtimes 100 (3)


Fdd Ffqd

Fq d

times 100 (4)










Secondaryindicators





Secondaryindicators





and strategy



of the supply chain






Cost profit margin




Profit growth rate

Human capital ratio





node

Sales profit margin


Inventory turnover












2 4 6 8 10 12 14

120E-05

100E-05

800E-06

600E-06

400E-06

200E-06

000E+00

200

180

160

140

120

100

80

60

40

20

0

Epoc

h

MSE

MSEEpoch



140

120

100

80

60

40

20

0

Time

Inve

ntor




Business center2502252001751501251007550250

Logi

stics

cost

()

600

500

400

300

200

100

000



Art

ifici

al co

st (

)

Business center

600

500

400

300

200

100

0002502252001751501251007550250







wij 1113944

m

k1V

ki V

kj ine j

0 i j

⎧⎪⎪⎨

⎪⎪⎩(5)




n (6)


Vi(t + 1) sgn 1113944n

j1wijVj(t)⎛⎝ ⎞⎠ (7)

Plan

Purchase

Technology

Produce

SalesCoordination


w11

w12

w13

w31

w32

w33

w21 w23w22

x1

y1 y2 y3

x2 x3

Layer 0

Layer 1




Vi(t + 1) Vj(t) (8)



system take A

0 15 051 0 12 0 08

⎡⎢⎢⎢⎢⎢⎣⎤⎥⎥⎥⎥⎥⎦ B

011

⎡⎢⎢⎢⎢⎢⎣⎤⎥⎥⎥⎥⎥⎦ Q

1 0 00 1 00 0 1

⎡⎢⎢⎢⎢⎢⎣⎤⎥⎥⎥⎥⎥⎦







y2520151050

x

08

06

04

02

0

-02

-04

X1X2X3


0 2 4 6 8 10 12 14 16 18 20

05

0

-05

-1

y

u


Time index

Fitti

ng ac

cura

cy in

dex

360

330

300

270310280250220190160




6 Conclusions



Data Availability




Acknowledgments


References













Time index310280250220190160

Fitti

ng ac

cura

cy in

dex

360

350

340

330

320

310

300

390

280





































Tpd 11139444

i1ti (1)






Rmc 1113936

mj1 F

jiq

1113936mj1 F

jtiq

times 100 (2)





Performance plan



Performance results




Rrud Ffud

Fudtimes 100 (3)


Fdd Ffqd

Fq d

times 100 (4)










Secondaryindicators





Secondaryindicators





and strategy



of the supply chain






Cost profit margin




Profit growth rate

Human capital ratio





node

Sales profit margin


Inventory turnover












2 4 6 8 10 12 14

120E-05

100E-05

800E-06

600E-06

400E-06

200E-06

000E+00

200

180

160

140

120

100

80

60

40

20

0

Epoc

h

MSE

MSEEpoch



140

120

100

80

60

40

20

0

Time

Inve

ntor




Business center2502252001751501251007550250

Logi

stics

cost

()

600

500

400

300

200

100

000



Art

ifici

al co

st (

)

Business center

600

500

400

300

200

100

0002502252001751501251007550250







wij 1113944

m

k1V

ki V

kj ine j

0 i j

⎧⎪⎪⎨

⎪⎪⎩(5)




n (6)


Vi(t + 1) sgn 1113944n

j1wijVj(t)⎛⎝ ⎞⎠ (7)

Plan

Purchase

Technology

Produce

SalesCoordination


w11

w12

w13

w31

w32

w33

w21 w23w22

x1

y1 y2 y3

x2 x3

Layer 0

Layer 1




Vi(t + 1) Vj(t) (8)



system take A

0 15 051 0 12 0 08

⎡⎢⎢⎢⎢⎢⎣⎤⎥⎥⎥⎥⎥⎦ B

011

⎡⎢⎢⎢⎢⎢⎣⎤⎥⎥⎥⎥⎥⎦ Q

1 0 00 1 00 0 1

⎡⎢⎢⎢⎢⎢⎣⎤⎥⎥⎥⎥⎥⎦







y2520151050

x

08

06

04

02

0

-02

-04

X1X2X3


0 2 4 6 8 10 12 14 16 18 20

05

0

-05

-1

y

u


Time index

Fitti

ng ac

cura

cy in

dex

360

330

300

270310280250220190160




6 Conclusions



Data Availability




Acknowledgments


References













Time index310280250220190160

Fitti

ng ac

cura

cy in

dex

360

350

340

330

320

310

300

390

280






























Tpd 11139444

i1ti (1)






Rmc 1113936

mj1 F

jiq

1113936mj1 F

jtiq

times 100 (2)





Performance plan



Performance results




Rrud Ffud

Fudtimes 100 (3)


Fdd Ffqd

Fq d

times 100 (4)










Secondaryindicators





Secondaryindicators





and strategy



of the supply chain






Cost profit margin




Profit growth rate

Human capital ratio





node

Sales profit margin


Inventory turnover












2 4 6 8 10 12 14

120E-05

100E-05

800E-06

600E-06

400E-06

200E-06

000E+00

200

180

160

140

120

100

80

60

40

20

0

Epoc

h

MSE

MSEEpoch



140

120

100

80

60

40

20

0

Time

Inve

ntor




Business center2502252001751501251007550250

Logi

stics

cost

()

600

500

400

300

200

100

000



Art

ifici

al co

st (

)

Business center

600

500

400

300

200

100

0002502252001751501251007550250







wij 1113944

m

k1V

ki V

kj ine j

0 i j

⎧⎪⎪⎨

⎪⎪⎩(5)




n (6)


Vi(t + 1) sgn 1113944n

j1wijVj(t)⎛⎝ ⎞⎠ (7)

Plan

Purchase

Technology

Produce

SalesCoordination


w11

w12

w13

w31

w32

w33

w21 w23w22

x1

y1 y2 y3

x2 x3

Layer 0

Layer 1




Vi(t + 1) Vj(t) (8)



system take A

0 15 051 0 12 0 08

⎡⎢⎢⎢⎢⎢⎣⎤⎥⎥⎥⎥⎥⎦ B

011

⎡⎢⎢⎢⎢⎢⎣⎤⎥⎥⎥⎥⎥⎦ Q

1 0 00 1 00 0 1

⎡⎢⎢⎢⎢⎢⎣⎤⎥⎥⎥⎥⎥⎦







y2520151050

x

08

06

04

02

0

-02

-04

X1X2X3


0 2 4 6 8 10 12 14 16 18 20

05

0

-05

-1

y

u


Time index

Fitti

ng ac

cura

cy in

dex

360

330

300

270310280250220190160




6 Conclusions



Data Availability




Acknowledgments


References













Time index310280250220190160

Fitti

ng ac

cura

cy in

dex

360

350

340

330

320

310

300

390

280

























Rrud Ffud

Fudtimes 100 (3)


Fdd Ffqd

Fq d

times 100 (4)










Secondaryindicators





Secondaryindicators





and strategy



of the supply chain






Cost profit margin




Profit growth rate

Human capital ratio





node

Sales profit margin


Inventory turnover












2 4 6 8 10 12 14

120E-05

100E-05

800E-06

600E-06

400E-06

200E-06

000E+00

200

180

160

140

120

100

80

60

40

20

0

Epoc

h

MSE

MSEEpoch



140

120

100

80

60

40

20

0

Time

Inve

ntor




Business center2502252001751501251007550250

Logi

stics

cost

()

600

500

400

300

200

100

000



Art

ifici

al co

st (

)

Business center

600

500

400

300

200

100

0002502252001751501251007550250







wij 1113944

m

k1V

ki V

kj ine j

0 i j

⎧⎪⎪⎨

⎪⎪⎩(5)




n (6)


Vi(t + 1) sgn 1113944n

j1wijVj(t)⎛⎝ ⎞⎠ (7)

Plan

Purchase

Technology

Produce

SalesCoordination


w11

w12

w13

w31

w32

w33

w21 w23w22

x1

y1 y2 y3

x2 x3

Layer 0

Layer 1




Vi(t + 1) Vj(t) (8)



system take A

0 15 051 0 12 0 08

⎡⎢⎢⎢⎢⎢⎣⎤⎥⎥⎥⎥⎥⎦ B

011

⎡⎢⎢⎢⎢⎢⎣⎤⎥⎥⎥⎥⎥⎦ Q

1 0 00 1 00 0 1

⎡⎢⎢⎢⎢⎢⎣⎤⎥⎥⎥⎥⎥⎦







y2520151050

x

08

06

04

02

0

-02

-04

X1X2X3


0 2 4 6 8 10 12 14 16 18 20

05

0

-05

-1

y

u


Time index

Fitti

ng ac

cura

cy in

dex

360

330

300

270310280250220190160




6 Conclusions



Data Availability




Acknowledgments


References













Time index310280250220190160

Fitti

ng ac

cura

cy in

dex

360

350

340

330

320

310

300

390

280

























and strategy



of the supply chain






Cost profit margin




Profit growth rate

Human capital ratio





node

Sales profit margin


Inventory turnover












2 4 6 8 10 12 14

120E-05

100E-05

800E-06

600E-06

400E-06

200E-06

000E+00

200

180

160

140

120

100

80

60

40

20

0

Epoc

h

MSE

MSEEpoch



140

120

100

80

60

40

20

0

Time

Inve

ntor




Business center2502252001751501251007550250

Logi

stics

cost

()

600

500

400

300

200

100

000



Art

ifici

al co

st (

)

Business center

600

500

400

300

200

100

0002502252001751501251007550250







wij 1113944

m

k1V

ki V

kj ine j

0 i j

⎧⎪⎪⎨

⎪⎪⎩(5)




n (6)


Vi(t + 1) sgn 1113944n

j1wijVj(t)⎛⎝ ⎞⎠ (7)

Plan

Purchase

Technology

Produce

SalesCoordination


w11

w12

w13

w31

w32

w33

w21 w23w22

x1

y1 y2 y3

x2 x3

Layer 0

Layer 1




Vi(t + 1) Vj(t) (8)



system take A

0 15 051 0 12 0 08

⎡⎢⎢⎢⎢⎢⎣⎤⎥⎥⎥⎥⎥⎦ B

011

⎡⎢⎢⎢⎢⎢⎣⎤⎥⎥⎥⎥⎥⎦ Q

1 0 00 1 00 0 1

⎡⎢⎢⎢⎢⎢⎣⎤⎥⎥⎥⎥⎥⎦







y2520151050

x

08

06

04

02

0

-02

-04

X1X2X3


0 2 4 6 8 10 12 14 16 18 20

05

0

-05

-1

y

u


Time index

Fitti

ng ac

cura

cy in

dex

360

330

300

270310280250220190160




6 Conclusions



Data Availability




Acknowledgments


References













Time index310280250220190160

Fitti

ng ac

cura

cy in

dex

360

350

340

330

320

310

300

390

280






























2 4 6 8 10 12 14

120E-05

100E-05

800E-06

600E-06

400E-06

200E-06

000E+00

200

180

160

140

120

100

80

60

40

20

0

Epoc

h

MSE

MSEEpoch



140

120

100

80

60

40

20

0

Time

Inve

ntor




Business center2502252001751501251007550250

Logi

stics

cost

()

600

500

400

300

200

100

000



Art

ifici

al co

st (

)

Business center

600

500

400

300

200

100

0002502252001751501251007550250







wij 1113944

m

k1V

ki V

kj ine j

0 i j

⎧⎪⎪⎨

⎪⎪⎩(5)




n (6)


Vi(t + 1) sgn 1113944n

j1wijVj(t)⎛⎝ ⎞⎠ (7)

Plan

Purchase

Technology

Produce

SalesCoordination


w11

w12

w13

w31

w32

w33

w21 w23w22

x1

y1 y2 y3

x2 x3

Layer 0

Layer 1




Vi(t + 1) Vj(t) (8)



system take A

0 15 051 0 12 0 08

⎡⎢⎢⎢⎢⎢⎣⎤⎥⎥⎥⎥⎥⎦ B

011

⎡⎢⎢⎢⎢⎢⎣⎤⎥⎥⎥⎥⎥⎦ Q

1 0 00 1 00 0 1

⎡⎢⎢⎢⎢⎢⎣⎤⎥⎥⎥⎥⎥⎦







y2520151050

x

08

06

04

02

0

-02

-04

X1X2X3


0 2 4 6 8 10 12 14 16 18 20

05

0

-05

-1

y

u


Time index

Fitti

ng ac

cura

cy in

dex

360

330

300

270310280250220190160




6 Conclusions



Data Availability




Acknowledgments


References













Time index310280250220190160

Fitti

ng ac

cura

cy in

dex

360

350

340

330

320

310

300

390

280



























wij 1113944

m

k1V

ki V

kj ine j

0 i j

⎧⎪⎪⎨

⎪⎪⎩(5)




n (6)


Vi(t + 1) sgn 1113944n

j1wijVj(t)⎛⎝ ⎞⎠ (7)

Plan

Purchase

Technology

Produce

SalesCoordination


w11

w12

w13

w31

w32

w33

w21 w23w22

x1

y1 y2 y3

x2 x3

Layer 0

Layer 1




Vi(t + 1) Vj(t) (8)



system take A

0 15 051 0 12 0 08

⎡⎢⎢⎢⎢⎢⎣⎤⎥⎥⎥⎥⎥⎦ B

011

⎡⎢⎢⎢⎢⎢⎣⎤⎥⎥⎥⎥⎥⎦ Q

1 0 00 1 00 0 1

⎡⎢⎢⎢⎢⎢⎣⎤⎥⎥⎥⎥⎥⎦







y2520151050

x

08

06

04

02

0

-02

-04

X1X2X3


0 2 4 6 8 10 12 14 16 18 20

05

0

-05

-1

y

u


Time index

Fitti

ng ac

cura

cy in

dex

360

330

300

270310280250220190160




6 Conclusions



Data Availability




Acknowledgments


References













Time index310280250220190160

Fitti

ng ac

cura

cy in

dex

360

350

340

330

320

310

300

390

280

























Vi(t + 1) Vj(t) (8)



system take A

0 15 051 0 12 0 08

⎡⎢⎢⎢⎢⎢⎣⎤⎥⎥⎥⎥⎥⎦ B

011

⎡⎢⎢⎢⎢⎢⎣⎤⎥⎥⎥⎥⎥⎦ Q

1 0 00 1 00 0 1

⎡⎢⎢⎢⎢⎢⎣⎤⎥⎥⎥⎥⎥⎦







y2520151050

x

08

06

04

02

0

-02

-04

X1X2X3


0 2 4 6 8 10 12 14 16 18 20

05

0

-05

-1

y

u


Time index

Fitti

ng ac

cura

cy in

dex

360

330

300

270310280250220190160




6 Conclusions



Data Availability




Acknowledgments


References













Time index310280250220190160

Fitti

ng ac

cura

cy in

dex

360

350

340

330

320

310

300

390

280
























6 Conclusions



Data Availability




Acknowledgments


References













Time index310280250220190160

Fitti

ng ac

cura

cy in

dex

360

350

340

330

320

310

300

390

280












































Documents

Performance Evaluation of Enterprise Supply Chain