Construction of supply chain management information system based on networked web service composition technology

Abstract

Web service composition is crucial for creating valuable services by integrating pre-existing ones. With their service-oriented architecture (SOA), which can be used for any system design, web services can increase flexibility. Fusing Web services architecture with Semantic Web services can better assist supply chain coordination in a distributive, autonomous, and ever-changing corporate environment than current information technology. Decisions must be made quickly and with enough information many systems fail to provide real-time supply chain insight. Forecasting, inventory management, and decision-making may all be impacted by poor data quality. Modifying preexisting systems according to unique organizational needs may be challenging and expensive. Hence, this paper proposes a semantic web service-based supply chain management framework (SWS-SCMF) to analyze the web services in supply chains and examine how they interact using Web Ontology Language (OWL)-S, including automated discovery, construction, and invocation. The suggested method for improving supply chain data integration uses an ontology-based multiple-agent decision support system. Different accessibility tools, data formats, management information systems, semantic web, and databases are integrated across the five interconnected levels of the system. Businesses may find the proposed approach useful for data and information sharing when dealing with complex supply chain management. The suggested SWS-SCMF is an adaptable, accurate, and effective method for bidirectional chaining composition that uses mediators to enable the automated composition of Semantic Web services. The numerical results show that our proposed method enhances the overall performance ratio by 94%, accuracy ratio by 98%, and supply chain management ratio by 91% compared to other methods.

Keywords

Supply chain management (SCM)web service composition (WSC)Web Ontology Language (OWL)-S semantic web service architecture (SWSA)

1. Introduction of supply chain management based on web service composition technology

Supply Chain Management (SCM) is a crucial operational strategy for improving organizational competitiveness and responding quickly to changing market conditions [1]. SCM allows a supply chain (SC) of geographically dispersed businesses in many industries to effectively convert raw materials into finished goods and deliver those goods and services to clients when and where they need them [2]. SCM’s overarching goal is to streamline critical business operations by connecting product and service providers with the people who make a difference – the consumers and other stakeholders benefitting from the goods and services they purchase [3]. Information systems are crucial to supply chain management. In the modern, global, competitive, and ever-changing corporate world, SCM relies heavily on cooperation between company partners [4]. The primary goal of supply chain coordination is to increase channel efficiency and guarantee “mutual success” between supply chain partners by influencing the actions of every individual entity in the supply chain [5].

By working together on demand forecasts and replenishment, companies may get access to up-to-date information, eliminate demand uncertainty, decrease logistics and shipping costs, shorten lead times, and boost customer service and profitability [6]. Coordinative relationships mutually benefit all parties involved; upstream suppliers gain from enhanced consumer input on quality, cost improvements, and shared savings [7]. With better products offered at reduced prices, end users also stand to gain. Lack of knowledge and coordination solutions, high levels of interconnection, decentralization, limited rationality, and unpredictability are some underlying causes that make supply chain coordination difficult [8].

Limitations that affect the supply chain’s effective functioning, efficiency, and performance are referred to as constraints in Supply Chain Management Information Systems (SCMIS). Limitations in transit options, storage space, and manufacturing. Processing times for orders, quality checks, and packing are all examples of operational procedures. Budgetary constraints constrain investment in infrastructure, tech, and human resources. Ensure the data used for decision-making is accurate and reliable across the supply chain. Following all applicable rules and regulations, both domestic and foreign, including those about trade tariffs and import/export restrictions. Supply chain performance optimization, cost reduction, and service level improvement depend on good management of these restrictions. By using sophisticated data analytics, real-time monitoring, and strategic planning, Supply Chain Management Information Systems are crucial for discovering, assessing, and reducing these restrictions.

Agent approaches have lately exploded in popularity because of their many desirable traits, such as continuous execution, awareness of both the environment and agents inside it, independence, intelligence, adaptability, mobility, anthropomorphism, and reproduction [9]. Many fields use agents, including computer science, e-commerce, decision support systems, supply-chain management, etc. Most prior methods have focused on individual businesses with many agents and simulation systems [10].

Furthermore, they failed to consider semantics to provide flexible information inquiry utilizing various language phrases. A method for creating web services using agent mediation. In web-based environments, where things happen quickly and on a grand scale, business services are handed off to a swarm of autonomous problem solvers [11]. These agents can handle complicated tasks thanks to their constant situational awareness and ability to make judgments on the fly. In today’s computer world, agents can adapt to different situations by overcoming limitations [12].

The complexity of web service creation makes it impossible for humans to handle the process manually. Two main parts make up this process: planning and scheduling [13].

The planning step achieves the aim by automatically generating or predefining a composite service, and the actual web services are sought, chosen, and bound in the scheduling stage to implement the plan [14]. Web service standards and technology have advanced, so supply chains and a distributed computing infrastructure provide access to many web services. Simultaneously, the Semantic Web may provide the missing semantic interoperability for the Web services architecture [15]. So, many companies are considering Semantic Web services as a potential solution. These services integrate the expanding Web services architecture with the Semantic Web, allowing for better coordination and ultimately more successful partnerships with commercial partners.

Figure 1.

Web Service Composition (WSC).

In other words, middleware is responsible for converting service requests into data source queries. Middleware contains connections among service providers on the web. Middleware can derive matching web service outputs based on the requested inputs. After that, it adds individual Web services and the outputs of those Web services to the respective set while simultaneously adding the outputs of Web services to the inputs set. Repeat the procedures discussed above until the desired results of the query are obtained.

On the other hand, the requests are reflected and transmitted by supply chain ontology and OWL-S. Web service composition starts with Web service modelling, which characteristic binding or production rules may accomplish.

For this reason, the composition scheme may be automatically developed and optimized with the assistance of the service composition algorithm. In conclusion, the execution engine should be used to carry out the composition scheme (the composition framework illustration may be seen in Fig. 1). Distributors, wholesalers, manufacturers, and retailers are all links in a supply chain. The high integration cost and impracticality of dynamically integrating partners on demand are consequences of traditional supply chains that rely on tight and long-term integration with partners. Integrating supply chain partners in real-time and at a reasonable cost may be possible with the rise of web-based technology.

Many more companies offer services and solutions in e-supply chains built using web service composition than in conventional supply chains, and partners work together in real-time and dynamic ways. Achieving success in an e-supply chain relies on the capacity to quickly and dynamically choose the most suitable partners and efficiently coordinate the services along the chain.

Hence, this paper proposes a semantic web service-based supply chain management framework (SWS-SCMF) to analyze the web services in supply chains and examine how they interact using Web Ontology Language (OWL)-S, including automated discovery, construction, and invocation. The suggested method for improving supply chain data integration uses an ontology-based multiple-agent decision support system. Different accessibility tools, data formats, management information systems, semantic web, and databases are integrated across the five interconnected levels of the system. Businesses may find the proposed approach useful for data and information sharing when dealing with complex supply chain management. The suggested SWS-SCMF is an adaptable, accurate, and effective method for bidirectional chaining composition that uses mediators to enable the automated composition of Semantic Web services. In existing methods, the ability to effectively communicate and collaborate as supply chain partners is severely lacking in many systems. Lack of clear and concise communication may cause problems with comprehension, time lags, and overall efficiency in operations.

Supply Chain Management Information Systems (SCMIS) have greatly improved supply chain management using networked web service composition technologies. This innovation’s fresh features and advantages improve the supply chain’s responsiveness, efficiency, and flexibility. The supply chain can easily include previously incompatible systems and apps thanks to web service composition. Suppliers, manufacturers, distributors, and retailers are all involved in this process, so it’s important that they can interact and exchange data efficiently. Allows the supply chain to share data in real time, giving everyone access to the most up-to-date information. This enhances the ability to make decisions and adapt to market, supply, and demand changes.

The main contribution of this paper

Designing the suggested SWS-SCMF work for supply chain web service analysis, including automated discovery, development, and invocation using Web Ontology Language (OWL)-S.

ii)

A multi-agent decision-support system based on ontologies is proposed to enhance supply chain data integration.

iii)

The suggested SWS-SCMF is an adaptable, accurate, and effective method for bidirectional chaining composition that uses mediators to enable the automated composition of Semantic Web services.

iv)

The numerical results and the proposed SWS-SCMF have been performed to achieve a high-performance supply chain management and accuracy ratio compared to other methods.

The upcoming phase is as follows: Section 2 deliberates the literature review, Section 3 explores the proposed methodology, Section 4 discusses the numerical results and discussion, Section 5 concludes the overall paperwork, and Section 6 includes the challenges and future work.

2. Literature review

This study experimentally examined the competing values framework (CVF) approach for organizational culture (OC) on sustainable supply chain performance (SSCP) in the food manufacturing sectors of the UK and Greece [16]. The quantitative research and interviews with food business executives corroborated previous findings that all cultural factors, except rational culture, positively affected SSC. The suggested systems thinking framework records the relationship between the relevant structural components to investigate the underlying dynamics [17]. In sum, this study adds to the body of knowledge on supply chain risk management by bringing together the relationship between resilience and governance, classifying relevant types, and proposing research that policymakers and managers can use to guide their decisions. The conceptualization of a 7-element digital twin framework for supply chain management described by Ivanovo, D [18]. Innovation, people, leadership, organization, scope, task, and modelling are the seven main components of a digital twin in SCOM which stands for supply chain and operations management.

Blockchain-based collaboration framework (BCF) with smart contract-based resource sharing to improve supply chain management [19]. To illustrate how partners interact, the framework includes (a) the structure of the network, (b) principles for how the network should operate determined by the needs of supply working together, (c) a UML diagram to specify the sequence of interactions between smart contracts, and (d) an algorithm to validate and verify the smart contract networks. The plastics industry in developing nations, particularly India, can benefit from circular supply chain management (CSCM), as discussed in this article [20]. The obstacles are prioritized using a fuzzy analytical hierarchy process (AHP) technique. Managers and policymakers can use this study’s findings as a foundation for implementing CSCM effectively.

Nanda et al. [21] proposed the NAIBHSC method as a novel approach to integrating the Internet of Things (IoT) with blockchain technology in the healthcare supply chain. Plan-do-check-act (PDCA) based group decision-making model (PDCA-GD-MM) presents a new approach to assessing manufacturing companies’ GSCM performance [22]. This study uses a FTOPSIS technique to rank the companies and a fuzzy-analytical hierarchy process (FAHP) technique for calculating the weights of the requirements in an integrated fuzzy multi-criteria decision-making (MCDM) approach. A variety of SCRes strategies can be developed using an integrated Multi-criteria decision-making (MCDM) approach that is powered by AI-based algorithms, including Wavelet Neural Networks (WNN), Evaluation based on Length from Average Solution (EDAS), and Fuzzy systems [23]. The results demonstrate that the most effective methods for promoting SCRes tactics are agent-based systems, machine learning using big data, and fuzzy logic computing.

Table 1
Comparative analysis of existing methods

Number of methods	Method name	Key features	Advantages	Disadvantages
BCF [19]	Blockchain-based framework	UML sequence diagram	The demonstration architecture makes efficient network resource use possible, guaranteeing data authenticity and quality in supply networks.	Data components and the extent to which they are accessible and visible to certain players should also be recognized and separated, as data sharing across partners is a big worry owing to privacy concerns.
NAIBHSC [21]	Novel Approach for Integrated IoT (Internet of Things) With Blockchain in Health Supply Chain (NAIBHSC) approach	Smart Contract using blockchain and the solidity programming language for Ethereum’s public-permission	Our proposed NAIBHSC approach improves the smart health supply chain management system’s overall performance, reduces latency, and improves response time. For a group of 500 users, the average Transaction Per Second (TPS) is 100 milliseconds.	An effective vehicle identification system is necessary for smart transportation due to the strong cryptographic techniques needed to prevent unauthorized users from tampering with data and information.
PDCA-GD-MM [22]	Plan-do-check-act (PDCA) based group decision-making model (PDCA-GD-MM)	Creating a questionnaire, forming a committee of decision-makers, and administering a survey are the methods used to gather data	These studies provide a methodology for evaluating performance that considers all three aspects of sustainability: environmental, social, and operational. Thus, general sustainability is directly affected by the benchmark organization’s approach.	Errors may be present owing to an expert’s impression of preference; This research employs an integrated decision-making technique that appears to be highly accurate.
IMCDMA [24]	Integrated multi-criteria decision-making approach (IMCDMA)	A committee of specialists from the strategic, tactical, and operational levels is formed, and surveys are conducted to gather data	Policymakers may find this information useful in developing sustainable sourcing strategies.	Fewer data gathered in the study

The entropy method (EM) is suggested to determine the criteria weights in an integrated multi-criteria decision-making approach (IMCDMA). At the same time, grey relational analysis ranks the supplier organizations independently [24]. The manufacturing firm stands out as the standard; other businesses may learn from its techniques to improve their performance. One of the data-driven approaches is Reinforcement Learning, a collection of algorithms for machine learning [25]. Using a semi-systematic approach (SSA), this literature review classifies existing research on reinforcement learning concerning supply chain management (SCM). Blockchain-based collaborative framework (BCF) is a system that uses smart contracts to pool resources [26]. In the framework, there is a network architecture that shows how partners interact, rules for how the network should work based on the needs of supply collaboration, a UML diagram that shows the sequence of interactions between smart contracts, and an algorithm that validates and verifies the smart contract network.

An adequate supplier selection process necessitates the use of best-worst methods (BWM) and geometrical analysis for interactive assistance (GAIA) [27]. Five pharmaceutical sector vendors used this hybrid approach to achieve results. The primary need was for the product to be delivered on time.

The significance of circular supply chains (CSCs) in enhancing the circularity of materials, components, and products has garnered growing attention in the literature on supply chain management [28]. This article aims to assist practitioners in identifying and mitigating the risks and uncertainties that hinder the adoption of CSCs and associated management techniques. Table 1 shows the comparative analysis of existing method merits and demerits.

Based on our analysis, there are some drawbacks to the existing methods, such as decision-making and accuracy. The proposed SWS-SCMF with the ontology-based multiple-agent decision-support system to enhance the supply chain. An innovative SWS-SCMF method for managing supply chains has been introduced by deploying networked web service composition technology. Organizations may strengthen their supply chains and make them more robust by using interoperable web services to increase efficiency, responsiveness, and flexibility. In the ever-changing global supply chain management field, this technological advancement solves current problems and creates new opportunities for improvement and expansion.

3. Proposed methodology

Web Ontology Language (OWL)-S, which encompasses automated discovery, development, and invocation of Semantic web services, defines and analyses the interactions between various web services in supply chains. As a result, SWS-SCMF is proposed in this study. An ontology-based multiple-agent decision-support system is proposed to enhance data integration in the supply chain.

All five levels work together to include various systems for management information, databases, semantic webs, accessibility tools, and data formats. In the face of complicated supply chain management, businesses may find the suggested method helpful for exchanging data and information.

To facilitate the automated composition of Semantic Web services, the proposed SWS-SCMF provides a flexible, precise, and efficient approach to bidirectional chaining composition via mediators.

Semantic Web Services are being integrated into web services to improve the automation and intelligence of web service discovery, creation, and invocation. By using ontologies, vocabularies, and semantic annotations, SWS enhances conventional web services by offering more comprehensive, machine-readable descriptions of the services’ capabilities, interfaces, and semantics. Because of this, automated, semantically-based service discovery, construction, and invocation are possible with greater accuracy. Adding a semantic layer to web services allows for improved intelligence and automated interactions. In contrast, web service composition concentrates primarily on the technical aspects of organizing and integrating individual services to achieve a larger business process or task. Both web service composition and Semantic Web Service are connected by combining and using web services.

This part proposes a structure for the e-SCM multi-agent system. There are five levels to the suggested architecture: access, communication, application, ontology, and database. The connections between the five levels are shown in Fig. 2. Listed below are the primary roles that these layers play.

Figure 2.

Proposed SWS-SCMF.

Layer 1: First, there’s the access layer, which acts as the interface for the user. Several tools are available to users who want to use the e-SCM multi-agents system. Devices such as desktop computers, mobile phones, laptops, personal digital assistants, and many more are available to consumers. Accessibility is a key feature of the access layer. From any location, users may connect to the e-SCM multi-agents system.

Layer 2: At the second level, known as the communication layer, data formats are transformed so that the system and various interfaces can receive them. For this purpose, three conversion agents have been developed: one for online inputs, one for radio inputs, and one for mobile inputs. Figure 2 shows that each agent primarily consists of four modules: receiving, assessment, transformation, and response. The evaluation module determined whether or not to provide the service after receiving data from the interface, which was then processed by the receiving module. If the evaluation is successful, the transformation module modifies the user’s data before sending it to the e-SCM multi-agents system. The system’s response module then transmits it to the user interface after the transformation module converts the internal format to the output format if the system receives any outputs.

Layer 3: The third and central layer of the e-SCM multi-agent system is the application layer. The e-SCM layer is another name for it. Specifically, SCM, ERP, and CRM are the three systems that make up this layer. There are several agents in each system that work together to complete tasks. These systems and agents must communicate and collaborate effectively to achieve their full potential.

SCM Systems: Its job is to fix issues arising in a supply chain due to suppliers and purchases. In incoming logistics, it employs several agents to carry out supply planning, procurement, demand planning, and production scheduling activities.

ii)

ERP Systems: ERP Systems’ job is to fix issues that arise inside the company. The integration requirements become more complicated when dealing with upstream and downstream data and information in a supply chain, as opposed to SCM and CRM systems. Managing orders in real-time, creating a demand plan, scheduling manufacturing, and distributing goods are all handled by several agents.

iii)

CRM Systems: As part of a supply chain, CRM resolves consumer complaints. Its integration requirements are more challenging than SCM and CRM systems because it must manage data and information from supply chains upstream and downstream. A team of agents handles order handling, distribution, online selling, and customer care functions.

Layer 4: The fourth layer of the e-SCM multi-agent system is the ontology layer, which manages the ideas and terminology used therein. A set of terminology, their semantic links, and some basic logic and inference principles applicable to supply chains are included. The ontology layer will convert words not already used to ensure database consistency. In this layer, the system can demonstrate its semantic behaviours.

Layer 5: The fifth layer is the database layer, which helps many business activities by storing relevant data in supply chains. For the e-SCM multi-agent system, it is crucial. This layer contains a plethora of databases. Orders, clients, production, packages, tracking, suppliers, purchases, inventories, and so on might all have their database. In most cases, people will utilize databases for querying, maintaining, and mining data. For semantic consideration, they may be efficiently used via the ontology platform. It undergoes a multi-stage processing to prepare data for storage in a database.

3.1 Semantic web service

Figure 3.

Semantic web service.

Figure 3 shows the SWSA computational framework. The specifications for a new service, including its inputs, outputs, preconditions, and more, are submitted by service consumers to the suggested architecture. Web service composition descriptions that have been upgraded conceptually are entered by hand in the SWSA service repository. When processing ontological concept matching, the distance connection between ideas and inheritance relations is considered when addressing similarity between concepts. Using a semantic distance-based concept similarity matching approach, SWSA considers the inheritance relationships and the overall distance between ideas.

The SWSA uses structural case-based reasoning (S-CBR) 1 to store services and related ontological concepts.

The evaluation of service similarity employs rule-based reasoning (RBR). Figure 3 shows the algorithm in action as it searches SWSA for marketed semantic web services.

Algorithm 1: SWS-SCMF similarity computation
Inward: two set of concepts $({D_{1},D_{2}})$ , root node $\textit{roo}_{m}$ , concepts graph (h)
Outward: semantic similarity value between two sets of concepts
1. Begin
2. If $D_{1}$ and $D_{2}$ are the same concept then $\textit{sim}_{c}=0$
3. else
4. if $D_{1}$ and $D_{2}$ are directly connected then $\textit{sim}_{c}=z({D_{1},D_{2}})$
5. else
6. if a direct path connection exists then
7. $T_{\textit{path }01}=\textit{shortest path}({H,D_{1},\textit{root}_{m}})$
8. $T_{\textit{path }02}=\textit{shortest path}({H,D_{2},\textit{root}_{m}})$
9. $\textit{sim}_{c}=z({T_{\textit{path }01}})+z({T_{\textit{path }02}})-2\ast z({% SP})$
// SP denotes the common sub-smallest path
10. end if
11. $\sigma({D_{1},D_{2}})=\frac{1}{\textit{deg}\ast\textit{sim}_{c}({D_{1},D_{2}})% +1}$
12. end if
13. end if
14. return $\sigma$
15. end

The SWSA similarity measure takes a numerical value between 0 and 1 to indicate how similar two concepts are, $D_{j}$ and $D_{i}$ . Greater values indicate a stronger resemblance between the ideas and may be seen as a one-way link. Here is how the notion of similarity is explained:

$\displaystyle\forall D_{1}\in H,\sigma({D_{1},D_{1}})=1$ $\displaystyle\forall D_{1},D_{2}\in H,0\leqslant\sigma({D_{1},D_{2}})\leqslant 1$ (1) $\displaystyle\forall D_{1},D_{2},D_{3}\in H,\textit{IF sim}_{c}({D_{1},D_{2}})% >\textit{sim}_{c}({D_{1},D_{3}})\textit{ then }\sigma({D_{1},D_{2}})<\sigma({D% _{1},D_{3}})$

As shown in Eq. (3.1), the similarity function has been described. The semantic similarity function’s range is provided by the attributes above $\sigma({D_{j},D_{i}})$ . When two concepts are conceptually identical say that they are similar $\sigma({D_{j},D_{i}})=1$ ; when they are completely different say that they are similar $\sigma({D_{1},D_{2}})=0$ . The result of the similarity function should fall on the closed interval $[{0,1}]$ in this case. $D_{1}$ , $D_{2}$ , and $D_{3}$ stand for the three ideas in graph $H$ , and here indicates the semantic distance. The computing objective of SWSA has been served by the following semantic similarity ( $\sigma$ ) function:

$\displaystyle\sigma({D_{1},D_{2}})=\frac{1}{\textit{deg}\ast\textit{sim}_{c}({% D_{1},D_{2}})+1}$ (2)

As calculated in Eq. (2) semantic similarity function has been deliberated. In this case, $D_{1}$ and $D_{2}$ stand for two concepts, and ‘deg’ is the effect degree of semantic distance on semantic similarity; the value of deg should range from 0 to 1. To determine how similar two ideas are semantically utilize a weight allocation function, as seen below:

$\displaystyle Z({D_{n},D_{m}})=\left[{\max({\textit{depth}({d_{n}})})+\frac{% \textit{order number}({d_{n}})}{\textit{SNodes}(H)+1}+1}\right]^{-1}$ (3)

As explored in Eq. (3) weight allocation function has been described. In this context, $\textit{SNodes}(H)$ and $\textit{order number}({d_{n}})$ denote the overall number of nodes in the concept graph $H$ and the order number of the node $D_{m}$ between their siblings, respectively. $D_{n}$ is the deepest possible level of a node; for example, the depth of a root node is 0, while the depth of a node immediately connected to it is 1.

A basic business sales environment is used to elucidate the features of the system that has been created. The idea is to have a mobile app connecting to an internet store so customers can buy jerseys. The process of matching a request with the advertised service for selling jackets is examined in the following instance.

An algorithm attempts to do semantic matching to choose the most appropriate coat. Figure 3 depicts the algorithm in action, which receives as inputs the root node $\textit{root}_{m}$ and the ideas graph ( $h$ ), and produces an output indicating the degree to which the two concepts are semantically comparable. Every node in this hierarchy stands for a different idea. The experimental comparison considers the semantic similarities of jumpers, pullovers, waistcoats, sweaters, vests, and cardigans.

3.2 Web service composition and semantic web service collaboration

Figure 4.

Collaboration of web service composition and semantic web service.

The linked level between integrated systems is lowered greatly by adopting the system-integrated approach based on the WSC model, and the system may be more extendable, reusable, and easier to maintain. Figure 4 shows the four steps of the WSC integrated system implementation process: setting up web services and ontology knowledge, the semantic process, choosing and combining web services, developing BPEL documents, and finally, implementing Business process execution language (BPEL).

Step 1: UDDI register centre registration of online services is a requirement for service providers. This is the only option for users to access the online services they need via the UDDI registration centre. Domain model parameters need ontology to eradicate the synthetic semantic issue.

Step 2: As a class, web services are considered. As characteristics, web services have input, output, and constraint options. Union semantics describes this data as a means of resolving semantic disputes. Eliminating semantic conflicts between input and output parameters is a prerequisite to combining web services to construct ontology for those parameters. Domain and web services model parameters are the components that makeup parameter ontology. The domain and web services models’ parameter mapping might be 1 to n or n to 1.

Step 3: Since SCM will include many online services, composition efficiency becomes crucial. A further factor that influences WSC is the synthetic algorithm and the semantic process, which have already been discussed. Most synthetic algorithms use semantic ontology-based forward and backward chain construction. Nevertheless, due to its inefficiency, the forward chain method tackles redundant web services and repeatedly mixes the same input and web services. The ordinary backward algorithm’s excessive backtracking further impacts the composition efficiency. There is an algorithm that can travel backwards without using backtracking. By loading closure to buffer at the beginning of the search, users may use this approach to increase the synthetic efficiency of web services.

Step 4: Based on the synthetic algorithm calculated earlier, a solution incorporating many web services is created and imported into the BPEL4WS code. The relevant engine will then execute this code to finish the composition function. By the web applications’ call orders, BPEL builder produces WSDL and BPEL documents that correspond to the synthetic web services. Massage, PortType, and Operation, Binding, Port, and Service are some of the generated WSDL documents. The BPEL document’s partner components provide the web services that the synthetic service uses, the variables specify the parameters that the synthetic service uses as inputs and outputs, and the sequence describes the procedure’s implementation structure.

Order fulfilment, inventory control, logistics management, and demand prediction are a handful of the supply chain operations that may be orchestrated by web service composition. The ability to compose web services makes integrating with business partners’ systems, such as wholesalers, retailers, manufacturers, and suppliers, easier. With SWS, businesses can automate finding and building supply chain services via semantic reasoning and matching. Organizations may dynamically create composite services to fulfil changing business needs by automatically discovering relevant services, assessing their compatibility and appropriateness for particular activities, and using semantic descriptions of services and their interactions. The combination of seamless integration, automated processes, and understanding made possible by Web service composition and Semantic Web Service greatly enhance the efficacy and efficiency of supply chain management. Companies may improve their supply chain operations, work together more effectively with their partners, and stay ahead of the competition in today’s fast-paced business world by using these technologies.

3.3 Web service composition-based Qos for SCM

Figure 5.

Web service composition based Qos for SCM.

The planned architecture for web service composition with a focus on quality of service is shown in Fig. 5. This design aims to develop an automated, user-friendly, quality-of-service (QoS) web service composition plan. An execution plan optimizer and a composition broker comprise the architecture shown in the picture. Optimizer for execution plans puts the suggested algorithm for web service composition with a focus on quality of service into action. The following is a simplified illustration of the execution steps:

The role of the composition broker is to assist the customer in identifying the desired composition schema. Accessing the composition schema from the schema repository is as easy as browsing a service category or keyword-based search. In addition, the composition broker uses the client-supplied weights for each of the six quality-of-service (QoS) attributes as inputs.

ii)

The composition broker uses the UDDI server to find possible web-based services for each transactional process in the chosen compositional schema. Then it downloads the WSDL-S documents from those WS providers.

iii)

To build a composition plan that complies with constraints, the strategy for execution optimizer performs the arrangement algorithm with the QoS values provided in the Web Services Description Language WSDL-S documents.

After the customer approves, the next step is to use an execution engine to implement the composition plan while keeping tabs on its progress. Building an OWL-S execution engine that can dynamically generate the service grounding is no simple feat, as mentioned in the beginning. With a pre-defined service grounding, it is feasible to process the OWL-S service model immediately.

There are both abstract and tangible components to WSDL’s characteristics. The web service requires semantic information, which includes attributes like the kind of message, the interface, and the actions that make up the interface. Binding, endpoints, and an application with a list of endpoints are attributes in the real-world component. The characteristics of the concrete component are used to access the online service hosted by the web service provider.

Web Services Description Language – Semantic is an enhanced version of WSDL that aims to provide additional semantic information to web services. The UDDI server is where web service providers enter their web service details. Figure 5 shows that UDDI stores the following nodes for each web service: business Entity, businessService, bindingTemplate, and tModel. Details about the service provider, like their name, address, phone number, etc., are included in the business entity. The web service’s name and endpoint URL are part of the business service. The access of tModel, which contains the WSDL-S document’s URL path, uses the bindingTemplate, a child node of the businessService. The WSDL-S document containing the values of the QoS properties may be accessed using the tModel in UDDI.

Using the client’s keyword and the website addresses of the services in a way that uses semantic linkages like hypernym and hyponyms may improve the effectiveness of the keyword-based search. Figure 5 shows the candidate service discovery module responsible for finding potential web services for each compositional schema atomic process.

4. Numerical results and discussion

In recent years, activity has been abundant in supply chain logistics information management research. Some are concentrating on finding a solution by using AI. Most of these methods typically consider a single company’s worth of agents and simulation systems. By the way, they hardly ever think about implementing semantics to enable flexible information querying using various language words. An SWS-SCMF for an e-SCM multiple-agent decision support system that incorporates ontology was presented to streamline the integration of data and information in supply chains. Access, communication, application, ontology, and database are the five levels that make up the system. Databases, management information systems, semantic web, data formats, and access tools are all interconnected across these levels. To accomplish the goal of supply chain integration and communication with minimal human involvement, several agents carry out distinct duties at each tier. The open and honest way in which companies interact is key to strategy. Companies involved in complex supply chain management may find it easier to share data and information using the proposed approach.

4.1 Experimental setup

The suggested web service composition algorithm’s performance was evaluated via comparative tests. The performance metric determines how long it takes for the computer to execute the plan to ensure that it complies with the constraints. System load, network circumstances, and hardware performance might cause these timeframes to fluctuate, take them as an indication only. Comprehensive performance profiling and load testing under different operating situations will provide more accurate and thorough performance measurements. Make that SCMIS works well and meets performance requirements in different environments by monitoring and improving execution times. Table 2 shows the experimental setup.

Table 2
Experimental setup

Parameter	Value
Processor	Intel Core i7 or AMD Ryzen 7
RAM	32 GB
Storage	1 TB SSD
Service provider	Google, AWS, Cloud or Azure
Number of devices	10, 50, 100
Development tool	Node.js is v20.6.0
Year of version	September 2023
Router	TP-Link Archer A7 AC1750
Switch	TP-Link TL-SG108 8-Port Gigabit Ethernet Network Switch
Operating system	Linux (Ubuntu)
Database	MongoDB (NoSQL)
Requirement analysis and planning	1–2 weeks
System design	2–4 weeks
Development	8–12 weeks
Integration and testing	4–6 weeks
Deployment	1–2 weeks
Monitoring and maintenance	9 days
Total duration for basic implementation	35 days
Load testing duration	30 minutes to 1 hour
Execution time	0.02 seconds
Check inventory operation	Approximately 0.0015 seconds
Update inventory operation	Approximately 0.0028 seconds
Process order operation	Approximately 0.015099 seconds

4.2 Dataset description

The investigation was conducted using data from Supply Chains, which DataCo Global uses. The Supply Chain dataset may use machine learning algorithms and R software. Registration focuses on the following areas: production, sales, provisioning, and commercial distribution. It also allows for generating new knowledge by correlating organized and unstructured data. Supply chain management is all about data collection, analysis, and interpretation of the flow of goods and services from producers to consumers. Purchases, goods, inventories, vendors, transportation, and customer demand are all part of the supply chain processes covered by the dataset. Through logistics optimization, inventory management, and predictive analytics, it seeks to maximize performance and efficiency in the supply chain. The dataset used for this study, “DataCo Smart Supply Chain for Big Data Analysis,” is publicly available for download. You can access it using the following link: https://drive.google.com/file/d/1vlUoaM75LJ12_RK6xwsqcIXg1VzZXzCR/view?usp= drive_link [29]. Table 3 dataset features explained.

Table 3
Dataset features analysis

Dataset	Description
Number of features	53
Days for shipping (real)	Type of transaction made
Days for shipment (scheduled)	Days of scheduled delivery of the purchased product
Benefit per order	Earnings per order placed
Sales per customer	Total sales per customer made per customer
Delivery status	Delivery status of orders: Advance shipping, Late delivery, Shipping cancelled, Shipping
	on time
Category name	Description of the product category
Customer password	Masked customer key
Customer email	Customer’s email
Department id	Customer ID
Market	Market to where the order is delivered: Africa, Europe, LATAM, Pacific Asia, USCA
Order id	Order code
Order status	COMPLETE, PENDING, CLOSED, PENDING_PAYMENT, CANCELED,
	PROCESSING, SUSPECTED_FRAUD, ON_HOLD, PAYMENT_REVIEW
Order region	Region of the world where the order is delivered: Southeast Asia, South Asia, Oceania,
	Eastern Asia, West Asia, West of USA, US Center, West Africa, Central Africa, North
	Africa, Western Europe, Northern, Caribbean, South America, East Africa, Southern
	Europe, East of USA, Canada, Southern Africa, Central Asia, Europe, Central America,
	Eastern Europe, South of USA
Data type	4 (Transfer, Debit, Cash, Payment)

i) Performance ratio (%)

Figure 6.

Performance ratio (%).

Several elements and variables must be considered to depict the effect of semantic web services on the efficiency of the supply chain. Supply chain systems can interact with one another without any problems since SWS uses common data standards and ontologies. This eliminates data silos and makes sure that data is interpreted similarly across all platforms. An equation that is simplified to show this connection is this:

$\displaystyle\textit{SCMP}=F+I+A+D+T$ (4)

Where SCMP denotes the supply chain management performance, $F$ expresses the Increased productivity as a result of better automation and interoperability, $I$ represents the improved decision-making capacities have an impact, $A$ discusses the quickness achieved by being able to adjust and adapt quickly, $D$ explains the influence of enhanced data transparency and audibility, $T$ signifies the strengthening of supply chain partner trust and openness. Equation (4) and Fig. 6 discuss supply chain management performance based on semantic web service. Various apps, platforms, and systems in the supply chain may integrate and work together seamlessly with the help of semantic web services. Regardless of language or data format barriers, supply chain participants may better comprehend and interact with one another when they use standardized semantic descriptions. Overall efficiency, accuracy, and process streamlining are all enhanced by this interoperability. With the help of semantic web services, it is possible to identify and compose services according to their semantic descriptions automatically. This allows SCM systems to integrate and discover relevant services on the go, allowing them to carry out intricate business procedures. Rich contextual information on the inputs, outputs, and capabilities of web services may be found in the semantic annotations linked with them.

Figure 7.

Precision ratio (%).

ii) Precision ratio (%)

As deliberated in Eq. (5) and Fig. 7 precision ratio has been expressed. Accuracy of data $A$ being handled correctly. Delayed data availability or data timeliness $T$ . A process’s level of automation determines how much room there is for human mistakes $L$ . Interoperability $I$ of systems. The capacity for various systems to work together and communicate without any problems. The scalability $S$ of a system is its capacity to take on ever-increasing data and transaction loads without sacrificing performance. Within the supply chain, web services allow for the smooth integration of heterogeneous systems across multiple enterprises. This interoperability makes accurate consolidation and processing of data from several sources possible. Online platforms enable supply chain participants to share data in real time. Decisions may be made with increased precision because of this real-time insight into inventory levels, shipping statuses, and manufacturing schedules.

$\displaystyle\textit{Precision}=f({A,T,I,S,L})$ (5)

iii) F1-score ratio (%)

Figure 8.

F1-score ratio (%).

As shown in Eq. (6), F1-score ratio. An accurate indicator of a system’s performance, the F1-score takes recall and precision into account. It becomes very beneficial when the importance of both false positives and false negatives is high or when there is a significant class distributional imbalance. Construct an equation that represents the effect of Web Service Composition Technology on Supply Chain Management Information System (SCMIS) recall and accuracy to enhance the F1 score. The SCMIS’s accuracy is in complete order precision (P). Use the SCMIS recall function (R) to find and complete all required orders.

$\displaystyle\textit{F1-score ratio}=\frac{P.R}{P+R}$ (6)

iv) Recall ratio (%)

Figure 9.

Recall ratio (%).

The recall ratio is discussed in Eq. (7) and Fig. 9. The recall is represented by $O$ . To change the scale and units proportionality constant $L$ . A data accuracy value, usually between 0 and 1 is represented by $B$ . $S$ is the data timeliness indicator; a smaller number implies a greater timeliness. The data timeliness is measured in seconds. An indicator of how well two systems work together, interoperability $J$ takes on a value between 0 and 1. Scalability, denoted by the letter $T$ and ranging from 0 to 1, indicates the system’s capacity to manage increasing requirements.

$\displaystyle\textit{Recall}=L.\frac{B.J.T}{S}$ (7)

v) Accuracy ratio (%)

Figure 10.

Accuracy ratio (%).

Automated data validation during integration and collection is possible with SWS using rules and logic specified in ontologies. As a result, errors may be found and fixed more easily, leading to higher-quality data. Using a uniform ontology to map various data schemas, SWS makes integrating data from diverse sources easier. This ensures that data from various systems are merged appropriately and decreases mismatches.

$\displaystyle\textit{Accuracy}=DC+DE+AQ+TE+CC$ (8)

As calculated in Eq. (8) and Fig. 10 the supply chain accuracy was examined. In this context, accuracy refers to Supply Chain Accuracy, $D C$ for the integration of data and uniformity, $D E$ for semantic Data enrichment, $A Q$ for automatic uses Validation and quality assurance, $T E$ for Tracking and audibility enhancements, and $C C$ for Collaboration and communication developments. Supply chain data may be enhanced with extra contextual information via semantic annotations linked to online services. By improving our comprehension of the data’s significance and correlations, this enhanced data makes decision-making more precise. When circumstances or demands in the supply chain change, semantic web services allow for dynamic adaptability. Supply chain management systems can automatically adapt workflows, data mappings, and processes to new circumstances without sacrificing accuracy because they use semantic descriptions of data and services. Data integration, enrichment, validation, dynamic adaption, traceability, collaboration, and communication are all made possible by semantic web services, which are vital in improving the precision of supply chain management. These features ensure that data used in supply chain activities is accurate and dependable, improving performance, efficiency, and decision-making.

vi) Similarity reduction analysis (%)

A semantic similarity metric compares two or more services’ inputs, outputs, functionalities, and restrictions. Machine learning algorithms trained on semantic representations, ontology-based techniques, and vector space models are some of the methods that may be used for semantic similarity computation. Use the selected similarity metric to determine the degree of semantic similarity between each pair of services in the supply chain. The outcome can be a similarity score between zero and one, with zero indicating a complete lack of resemblance and one indicating a complete identity of services. Table 4(a) denotes the synonymy similarity and (b) semantic similarity reduction based on an algorithm.

Table 4

(A) Synonymy similarity and (B) Synonymy similarity based on algorithm

(A) Synonymy similarity					(B) Semantic similarity based on algorithm
	$D_{1}$	$D_{2}$	$D_{3}$	$D_{4}$		$D_{1}$	$D_{2}$	$D_{3}$	$D_{4}$
$D_{1}$	1	0	0	0	$D_{1}$	1	0.50	0.47	0.48
$D_{2}$	0	1	0	0	$D_{2}$	0.50	1	0.42	0.47
$D_{3}$	0	0	1	0	$D_{3}$	0.44	0.56	1	0.52
$D_{4}$	0	0	0	1	$D_{4}$	0.45	0.54	0.60	1

Equations (3.1)–(3) calculated the semantic simulation computation and reduction. Set a cutoff point for how similar services are before merging or eliminating them if they are deemed redundant. Services that score higher than this similarity level are kept in the supply chain since they are deemed different. Use the performance assessment results and comments to refine the similarity reduction procedure iteratively. Supply chain terminology and data models may be standardized using SWS’s ontologies. Thanks to this standardization, the data will be identified as comparable regardless of its source or description. By modelling supply chain workflows using process ontologies, SWS may find duplicates or overlapping processes. Semantic analysis of these processes allows for simplifying or eliminating unnecessary phases, leading to increased efficiency.

vii) Supply chain management ratio (%)

Figure 11.

Supply chain management ratio (%).

SWS uses intelligent agents to negotiate, carry out, and oversee supply chain operations without human intervention. These agents can enhance the efficiency of procurement, logistics, and inventory management. Different system parts may be created, updated, and maintained separately according to SWS’s modular design philosophy. As a result, the supply chain management system has become more adaptable and scalable.

$\displaystyle\textit{SCM}=B+U+E+D+R$ (9)

As obtained in Eq. (9) and Fig. 11, supply chain management is examined. SCM stands for the efficacy of Supply Chain Management. Automating procedures via service composition is $B$ . The letter $U$ stands for the visibility that composed processes provide. Service composition enables flexibility and adaptation, represented by the letter $E$ . Composed workflows enable collaboration across supply chain participants, denoted by the letter $D$ . $R$ represents the best possible performance as a result of well-crafted service offerings. By facilitating the development of intricate business processes via integrating several separate online services, web service composition is essential to supply chain management (SCM). By composing web services, SCM systems may automate several supply chain operations, including logistics coordination, inventory management, and order processing.

viii) Decision making ratio (%)

Significant progress in supply chain management has been made by integrating networked web service composition technologies into Supply Chain Management Information Systems (SCMIS). Supply chains have become more efficient, adaptable, and quick to respond with the help of this innovation and its many useful features and advantages.

Figure 12.

Decision-making ratio (%).

Figure 12 deliberates the decision-making ratio (%). Guarantees that everyone in the supply chain can access the most up-to-date information by enabling real-time data interchange. Making better decisions and being more sensitive to developments in the market are both made easier by this. With web services, data can be easily integrated from many sources, giving decision-makers access to the most up-to-date and accurate information. It allows members to work together more effectively by giving them access to the same data. Predictive analytics and real-time data monitoring aid in the early detection of problems, enabling proactive ways to mitigate them.

4.3 Discussion

The execution of repeated operations may be automated by assembling individual services into bigger workflows, which reduces human effort and improves efficiency. This partnership improves activity synchronization and goal alignment because stakeholders can better communicate, coordinate, and work together. Supply chain management systems can maximize performance via service composition, which allows for the distribution and parallelization of work among various resources. By lowering latency and boosting throughput, among other benefits, this concurrent execution of activities may boost system performance, especially for computationally costly or time-critical operations. Table 5 examines the overall outcome of the results.

Table 5
Overall outcome of the results

Number of methods	BCF	NAIBHSC	PDCA-GD-MM	IMCDMA	SWS-SCMF
	[19]	[21]	[22]	[24]
Accuracy ratio (%)	70	65	75	61	98
Performance ratio (%)	80	66	76	61	94
Precision ratio (%)	68	62	61	73	90
F1-score ratio (%)	73	63	69	61	93
Recall ratio (%)	70	69	61	73	94
Supply chain management ratio (%)	75	80	61	65	91
Decision making ratio (%)	83	78	77	67	92

The null hypothesis is rejected since the $p$ -value (0.01) exceeds the significance threshold (0.05). According to the evidence, semantic web services greatly improve SCMIS’s prediction accuracy. Determine the confidence interval (usually 95% CI) around the effect size. The range within which the genuine difference in prediction accuracy resides with 95% confidence is provided [5.0%, 11.0%]. It’s possible to thoroughly assess how Semantic Web Services affect the precision of SSMIS by using $p$ -values, confidence intervals, and effect sizes. The proposed SWS-SCMF model enhances the overall performance ratio, accuracy ratio, and supply chain management ratio compared to other existing blockchain-based collaboration framework (BCF) [19], NAIBHSC [21], Plan-do-check-act (PDCA) based group decision-making model (PDCA-GD-MM) [22], integrated multi-criteria decision-making approach (IMCDMA) [24] models.

5. Endnotes

Semantic Web services, which combine Web services architecture with the Semantic Web, outperform state-of-the-art IT in facilitating supply chain coordination in decentralized, self-governing, inherently dynamic business settings. Therefore, this study suggests that SWS-SCMF catalogues various web services involved in supply chains and analyzes their interactions via Web Ontology Language (OWL)-S. This language encompasses automated discovery, development, and invocation of Semantic web services. An ontology-based multiple-agent decision-support system is proposed as a means to elevate the integration of supply chain data. Through its five interrelated layers, the system integrates various accessibility technologies, data formats, management information systems, the semantic web, and databases. When tackling complicated supply chain management, businesses may find the suggested method helpful for exchanging data and information as needed. The proposed SWS-SCMF is a flexible, precise, and efficient strategy for bidirectional chaining composition that uses mediators to facilitate the automated creation of Semantic Web services. Supply chain management information systems using Semantic Web Services provide greater data integration, more effective automation, scalability, flexibility, and collaboration. These enhancements lead to more adaptable and efficient supply chain operations, boosting overall performance and competitiveness. By analyzing the interdependencies and linkages between various supply chain operations, SWS can automate complicated workflows. Operations are sped up, and manual intervention is reduced. The numerical results show that our proposed method enhances the overall performance ratio by 94%, accuracy ratio by 98%, and supply chain management ratio by 91% compared to other methods.

6. Challenges and future research

This study will build upon the ontology-based reasoning architecture in the future by exploring other mapping strategies that better describe services. This will allow for deeper service discovery. Improvements in dynamic service discovery, semantic interoperability, SOA reliability, safety, tracking of performance, and integration of AI, ML, and blockchain technology should be the prime focus of future research on Supply Chain Management Information Systems built on networked Web Service Composition Technology. Supply chain management, industry-specific solutions, collaborative research, standardization initiatives, and practical implementations must be implemented to change this strategy.

Accurate provenance tracing in intricate supply networks is not always easy to achieve. The credibility and dependability of the information systems used for supply chain management might be damaged by tracking errors.

References

Echefaj

Charkaoui

Cherrafi

Tiwari

Sharma

Jabbour

CJC

. From linear to circular sustainable supply chain network optimization: Towards a conceptual framework. Production Planning & Control. 2024; Jan(13): 1-25.

Donyavi

Flanagan

Assadi-Langroudi

Parisi

. Understanding the complexity of materials procurement in construction projects to build a conceptual framework influencing supply chain management of MSMEs. International Journal of Construction Management. 2024; 24(2): 177-186.

Ruzo-Sanmartín

Abousamra

Otero-Neira

Svensson

. The role of supply chain integration between integrated information technology and financial performance – a disaggregated framework and findings. The International Journal of Logistics Management. 2024; 35(2): 483-503.

Chen

Cheng

Zhang

Chen

Wang

. Decarbonization practices in the textile supply chain: Towards an integrated conceptual framework. Journal of Cleaner Production. 2024; 435: 140452.

Hosseinnia Shavaki

Ebrahimi Ghahnavieh

. Applications of deep learning into supply chain management: A systematic literature review and a framework for future research. Artificial Intelligence Review. 2023; 56(5): 4447-4489.

Liu

Song

Liu

. Leveraging digital capabilities toward a circular economy: Reinforcing sustainable supply chain management with Industry 4.0 technologies. Computers & Industrial Engineering. 2023; 178: 109113.

Amini

Jahanbakhsh Javid

. A multi-perspective framework established on the diffusion of innovation (DOI) theory and technology, organization, and environment (TOE) framework toward supply chain management system based on cloud computing technology for small and medium enterprises. Organization and Environment (TOE) Framework Toward Supply Chain Management System Based on Cloud Computing Technology for Small and Medium Enterprises (January 2023). International Journal of Information Technology and Innovation Adoption. 2023; 11: 1217-1234.

Nezafat Tabalvandani

Hosseini Shirvani

Motameni

. Reliability-aware web service composition with cost minimization perspective: A multi-objective particle swarm optimization model in multi-cloud scenarios. Soft Computing. 2024; 28(6): 5173-5196.

De Giacomo

Favorito

Leotta

Mecella

Silo

. Digital twins composition in smart manufacturing via Markov decision processes. Computers in Industry. 2023; 149: 103916.

10.

Fujishima

. Knowledge-driven automated service composition as a method for developing decision support systems. International Journal for Applied Information Management. 2022; 2(1): 44-49.

11.

Raja Santhi

Muthuswamy

. Influence of blockchain technology in manufacturing supply chain and logistics. Logistics. 2022; 6(1): 15.

12.

Abdirad

Krishnan

. Examining the impact of E-supply chain on service quality and customer satisfaction: A case study. International Journal of Quality and Service Sciences. 2022; 14(2): 274-290.

13.

Shamout

Ben-Abdallah

Alshurideh

Alzoubi

Kurdi

Hamadneh

. A conceptual model for the adoption of autonomous robots in the supply chain and logistics industry. Uncertain Supply Chain Management. 2022; 10(2): 577-592.

14.

Montag

. Circular economy and supply chains: Definitions, conceptualizations, and research agenda of the circular supply chain framework. Circular Economy and Sustainability. 2023; 3(1): 35-75.

15.

Sethi

Chung

Choi

. Reforming global supply chain management under pandemics: The GREAT-3Rs framework. Production and Operations Management. 2023; 32(2): 524-546.

16.

Osei

Papadopoulos

Acquaye

Stamati

. Improving sustainable supply chain performance through organizational culture: A competing values framework approach. Journal of Purchasing and Supply Management. 2023; 29(2): 100821.

17.

Tsolakis

Zissis

Tjahjono

. Scrutinizing the interplay between governance and resilience in supply chain management: A systems thinking framework. European Management Journal. 2023; 41(1): 164-180.

18.

Ivanov

. Conceptualization of a 7-element digital twin framework in supply chain and operations management. International Journal of Production Research. 2024; 62(6): 2220-2232.

19.

Agrawal

Angelis

Khilji

Kalaiarasan

Wiktorsson

. Demonstration of a blockchain-based framework using smart contracts for supply chain collaboration. International Journal of Production Research. 2023; 61(5): 1497-1516.

20.

Khandelwal

Barua

. Prioritizing circular supply chain management barriers using fuzzy AHP: Case of the Indian plastic industry. Global Business Review. 2024; 25(1): 232-251.

21.

Nanda

Panda

Dash

. Medical supply chain integrated with blockchain and IoT to track the logistics of medical products. Multimedia Tools and Applications. 2023; 82(21): 32917-32939.

22.

Ghosh

Mandal

Ray

. A PDCA based approach to evaluate green supply chain management performance under fuzzy environment. International Journal of Management Science and Engineering Management. 2023; 18(1): 1-15.

23.

Belhadi

Kamble

Fosso Wamba

Queiroz

. Building supply-chain resilience: An artificial intelligence-based technique and decision-making framework. International Journal of Production Research. 2022; 60(14): 4487-4507.

24.

Ghosh

Mandal

Ray

. Green supply chain management framework for supplier selection: An integrated multi-criteria decision-making approach. International Journal of Management Science and Engineering Management. 2022; 17(3): 205-219.

25.

Rolf

Jackson

Müller

Lang

Reggelin

Ivanov

. A review on reinforcement learning algorithms and applications in supply chain management. International Journal of Production Research. 2023; 61(20): 7151-7179.

26.

Agrawal

Angelis

Khilji

Kalaiarasan

Wiktorsson

. Demonstration of a blockchain-based framework using smart contracts for supply chain collaboration. International Journal of Production Research. 2023; 61(5): 1497-1516.

27.

Ishizaka

Khan

Kheybari

Zaman

. Supplier selection in closed loop pharma supply chain: A novel BWM-GAIA framework. Annals of Operations Research. 2023; 324(1): 13-36.

28.

De Lima

Seuring

. A Delphi study examining risk and uncertainty management in circular supply chains. International Journal of Production Economics. 2023; 258: 108810.

29.

https://drive.google.com/file/d/1vlUoaM75LJ12_RK6xwsqcIXg1VzZXzCR/view?usp=drive_link.

Construction of supply chain management information system based on networked web service composition technology

Abstract

Keywords

1. Introduction of supply chain management based on web service composition technology

Table 1 Comparative analysis of existing methods

4.1 Experimental setup

Table 2 Experimental setup

Table 3 Dataset features analysis

i) Performance ratio (%)

ii) Precision ratio (%)

iii) F1-score ratio (%)

iv) Recall ratio (%)

v) Accuracy ratio (%)

vi) Similarity reduction analysis (%)

vii) Supply chain management ratio (%)

viii) Decision making ratio (%)

Table 5 Overall outcome of the results

6. Challenges and future research

References

Table 1
Comparative analysis of existing methods

Table 2
Experimental setup

Table 3
Dataset features analysis

Table 5
Overall outcome of the results