Blockchain technology for projects: a multi-criteria decision matrix Weisheng Lu, Liupengfei Wu, and Fan Xue This is the peer-reviewed, post-print version of the paper: Lu, W., Wu, L., & Xue, F., (2022). Blockchain technology for projects: a multicriteria decision matrix. Project Management Journal, 53(1), 84-99. Doi: https://doi.org/10.1177/87569728211061780 This material is shared under the permission of PMJ and may be downloaded for personal use only. Any other use requires prior permission of PMJ. Abstract 5 This research aims to develop a multi-criteria decision matrix (MCDM) for project management (PM) professionals, which will support blockchain type selection, evaluate blockchain platforms, and plan blockchain systems. The MCDM is substantiated through a case study that includes a questionnaire and an illustrative example pertinent to the construction industry. It is discovered that in this study, consortium blockchain is superior in dealing with the characteristics of projects, and Hyperledger Fabric is chosen as the best applicable platform. In planning a blockchain-based PM system, PM professionals should consider user requirements such as network participants, principal transactions, communication channels, and smart contracts. 10 Keywords: Blockchain; Project; Multi-criteria decision-making; Decision matrix; Construction Introduction 15 20 25 Blockchain first emerged as a cryptocurrency transaction ledger and has now found applications in other spheres, including intellectual property and digital identity management (Perera et al., 2020), philanthropy (Lamba et al., 2019), democracy and governance (Diallo et al ., 2018), agriculture (Yang et al., 2020), health (Griggs et al., 2018), logistics and supply chain management (Kaijun et al., 2018), built environment provision (Mukne et al., 2019), education and human rights (Turkanović et al., 2018), energy and environment (Pop et al., 2018), and finance (Chen et al., 2017). A blockchain is a list of blocks, and each block contains a block header and a set of transactions (Hyperledger, 2020). Xue and Lu (2020) identify three essential components of blockchain technology, sometimes also simply called blockchain: cryptographic algorithms, distributed databases, and consensus mechanisms. Together, these components allow important transaction data to be collected, agreed upon, and distributed with minimal risk of alteration or tampering. Its widely propagated benefits, therefore, include enhanced security, improved traceability, increased efficiency, and reduced costs (Penzes et al., 2018). 1 30 35 40 45 50 Researchers and practitioners in the project world are also exploring this disruptive technology. They are encountering its different types, often inconsistently named but distinguished by privileges such as access rights, querying transactions, and submitting transactions to the ledger. Perera et al. (2020), for instance, propose three types of blockchain: (1) public, (2) private, and (3) consortium, and note different blockchain platforms, both commercial and open source. Projects may also be perceived in differing ways in the literature. A project may be viewed as a temporary organization (Turner & Müller, 2003), a unique setting beyond the dichotomous view of markets and hierarchies (Powell, 1990), an intrafirm endeavor or an interfirm coalition (Winch, 1989), or a network of stakeholders and their interactions (Lu et al., 2020). Projects emerged because people noticed that work related to film making, media production, construction, information and technology, and research and development (R&D), for example, can be better delivered if organized in such a way. In the case of blockchain selection, however, the uniqueness, temporariness, uncertainty, discontinuity, and one-off nature of projects present challenges. Unlike banking transactions, project transactions happen infrequently amongst limited parties. In addition, some transaction information is urged to be shared among project team members, while other information is strictly treated as a commercial secret. Recent studies have compared blockchains but few, if any, have provided a holistic method for their selection in a project setting. This research aims to develop a decision-support tool for project management (PM) professionals, helping them to select proper blockchain to enhance project value delivery. The rest of the paper is organized as follows. Next section is a literature review introducing blockchain. The following section reviews projects and their salient features in the context of blockchain. The subsequent section introduces our research methods. After that, the decision matrix is presented and substantiated using a case study. Finally, the discussion and conclusions are presented. Blockchain 55 60 65 Blockchain basics As introduced briefly above, blockchain combines cryptography, distributed databases, and consensus mechanisms (Xue and Lu, 2020). Cryptography has one main sub-component: hash algorithm, which ensures the immutability of the recorded transaction information (e.g., transaction personnel, time, location). Due to its emphasis on the security, blockchain is an excellent fit in institutional economics as economic infrastructure (Allen et al., 2018). Transaction records are connected as blocks in sequence, starting with the genesis block (Figure 1). Hashing sequentially interconnected blocks means that transaction data (e.g., Tx0 to Txn) is used as input to a hash function that converts the data into a fixed length string (Xue & Lu, 2020). The resulting hash value is unique for each input, so if someone changes transaction data, the corresponding hash value will also change immediately. Each block also contains the 2 previous block hash, ensuring that it cannot be changed without also changing the previous block (Li et al., 2021). 70 75 Figure 1. A blockchain (Adapted from Li et al., 2021) The distributed database of the blockchain, mainly composed of ledgers, is achieved through a decentralized network. A ledger is an accounting technique used to record when anything of value is transacted (Taghizadeh-Hesary & Yoshino, 2019). Peer nodes are network entities that maintain a copy of the ledger and/or run smart contracts to perform query/write operations on the ledger (Hyperledger, 2020). A smart contract is a program stored in a blockchain system wherein the outcome of any self-execution of the program is recorded on the distributed ledger (ISO, 2020). All nodes are related to each other in a planar rather than hierarchical, centralized topology, and this protects the network from a single point of failure. 80 85 90 Consensus is a protocol for verifying the facts on the fundamental datum of an economy: of identity, property, contract, and value, and regularly recording time (Davidson et al., 2018). There are many consensus algorithms. The most common one is proof of work (PoW), which is used to confirm transactions then initiate new blocks to the chain. A relatively energy-saving mechanism to save computing/mining power, proof of stake (PoS) assumes that participants with a greater stake (e.g., coins) display less opportunistic behavior, so have more opportunities to verify blocks. Stakeholders may also elect delegates to validate blocks, a method called delegated proof of stake (DPoS). Hyperledger Fabric 2.0 uses practical Byzantine fault tolerance (PBFT), a low-energy mechanism that requires three rounds of voting (Hyperledger, 2020). Other consensus algorithms are proof of authority (PoA) (Kaur & Oza, 2020), proof of authentication (PoAh) (Puthal et al., 2020), proof of conformance (PoC) (Zhang & Lin, 2018), 3 Kafka consensus (Mukne et al., 2019), and AlgoRand consensus (Zhang & Wang, 2019), while customized consensuses have also been observed (e.g., Kim & Yu, 2018). 95 Blockchain has a set of unique terminologies. In the case of PM professionals, these are unnecessarily accessible. Therefore, Xue and Lu (2020) have proposed a two-sided strategy, leaving the system R&D in the lower layers to blockchain professionals such as computer scientists, and the applications in the upper layers to users to solve their domain problems. PMs should nonetheless be involved in defining their requirements and in blockchain selection. 100 Available blockchain options 105 110 115 Various blockchain taxonomies can be found in the literature. Perera et al. (2020) summarizes popular platforms into public, private, and consortium categories (see Table 1). Public blockchain is accessible to the public for use. It is a system wherein control is distributed among the persons or organizations participating in the operation of the system. Therefore, it provides transparency and auditability, and it is difficult for hackers to manipulate data across the decentralized network (ISO, 2020). The consensus mechanism is driven by financial incentives, motivating all to improve the network. It is much work for block validators to verify transactions every day. As compensation for their efforts, after validators approve and add a new block of transactions to the blockchain, they will receive coins (e.g., Bitcoins). However, the securityoriented method may compromise privacy, scalability, and transaction efficiency (She et al., 2019). Private blockchain has just one owner organization, and only pre-authorized participants can perform particular activities (ISO, 2020). Due to its restrictive nature, it tends to have better privacy, scalability, and efficiency than the public blockchain (Lee et al., 2019). In addition, because the consensus is based on permissions, operating costs can be lower (Halloush et al., 2019). However, the controlled environment may reduce transparency, auditability, and resistance to hackers (Penzes et al., 2018). 120 125 Consortium blockchain operates under a selected set of nodes (She et al., 2019). It shares many advantages with private blockchain including access control, privacy, efficiency, and scalability. A subset of organizations can have dedicated communication channels and private data, or in some cases, all participants can have transparent information (Hyperledger, 2020). Consortium blockchain reduces the counterparty risk of private blockchain because it has more than one organization to manage the network but provides no incentive for parties to participate in the network. It provides customizability through governance structures. However, establishing welldefined regulatory rules and censorship systems may be difficult. 4 130 Some studies categorize blockchain as permissioned or permissionless, based on whether the blockchain has access control to ledgers. Users are pre-authorized to participate in a permissioned blockchain. Private and consortium blockchains are permissioned (Perera et al., 2020), while public blockchain is permissionless. 5 Table 1. Snapshot of popular blockchain platforms 134 135 Blockchain platforms Type Permission type Governance Focus domain Permissionless Permissionless Permissionless Year of implementation 2009 2015 2018 Bitcoin Ethereum EOS.IO Public Public Public NEO Source model Cryptocurrency Cross-industry Cross-industry Smart contract No Yes Yes Bitcoin developers Ethereum developers EOS.IO core arbitration forum Private Permissioned 2016 NEO holders & NEO foundation Smart Economy Yes Open source Hyperledger Sawtooth R3 Corda Private Consortium Permissioned Permissioned 2018 2016 Linux foundation Hedera hashgraph council Ark developers Linux foundation Cross-industry Cross-industry Yes Yes Cross-industry Cross-industry Yes Yes Open source Commercial Open source Open source Open source Ark Hyperledger Fabric Consortium Private/ Consortium Permissioned Permissioned 2017 2016 Note: Compiled from Turkanović et al. (2018), LeewayHertz (2019), Ark (2019), Hyperledger (2020), Perera et al. (2020) 6 Open source Open source Open source & Projects 140 145 Perceptions of projects vary. The Project Management Institute (2000) considers a project to have a production function and sees it as a temporary endeavor. Turner and Müller (2003) add that it is an agency that manages uncertainty and assigns resources. Winch (1989) differentiates projects as an intrafirm endeavor or interfirm coalition. Following the central tenet of Allen (1984) that production organizations, projects are not only technical systems but also social systems of personal and group interactions. Wang et al. (2018b) and Lu et al. (2020) view the project as a social network, using Engwall’s (2003) proposition that “no project is an island” to emphasize projects’ social embeddedness. DeFillippi and Sydow (2016) and Sydow et al. (2016) perceive projects as networks at two different levels. From a PM perspective, the network is a single interorganizational project consisting of temporal and internal relationships facilitating its coordination and ensuring its foreseeable delivery. From a business perspective, the network is a series of projects that are interrelated through interorganizational relationships. In this study, we focus on the former. 150 155 160 165 170 A project has several key characteristics differentiating it from “business as usual” or production. Firstly, it has a unique purpose to do things that have not been done previously (Smyth, 2018). Secondly, a project is temporary, involving a project team and broader coalition temporarily coming together (Turner & Müller, 2003). Thirdly, it involves uncertainties since it creates change in a changing environment (Lu et al., 2020). In the early stages of project in particular, it is difficult to define the objectives, determine the required funds, and evaluate the completion time. Fourthly, a project is one-off, with a definite start and end (Wright & Kersner, 2004). Finally, a project suffers discontinuity (Smyth, 2018). Project members are often established and dispersed, and human resources allocated to them change constantly. These characteristics pose considerable challenges for blockchain selection. Project uniqueness requires customization of each blockchain solution according to scope of use and degree of network centralization. The temporariness and one-off nature of projects imply that the cost of blockchain should not be high and appropriately distributed to the project members. Uncertainties subject a project to opportunistic behavior, so project-oriented blockchain solutions must enhance auditability, build trust and improve transparency and traceability. Discontinuity may lead to file tampering and loss and resource changes, so blockchain solutions must protect privacy and security of documents. Finally, blockchain should remain scalable so that it can continue to be used as the project progresses. These challenges imply that the project, as a special organizational form, should have a preference for a specific type of blockchain. To the best of our knowledge, no previous study has 7 consciously considered which blockchain is preferential in a project setting with its various characteristics. 175 Research methods 180 185 190 195 200 205 210 To develop a blockchain decision-support tool for PM professionals, this research adopts a methodology comprising literature review, decision matrix method (DMM), and case study. The critical literature review was to collect criteria for selecting type of blockchain, blockchain platform, and planning a blockchain system in various industries or settings. We searched Google Scholar using the keywords “public blockchain”, “private blockchain”, “consortium blockchain”, and “block chain”. The search initially produced 1570 hits including journal and conference papers, books, dissertations, and reports. The titles and abstracts were screened for suitability. The hits that do not have a specific application but a general review was excluded. The full texts of 214 papers passing the primary screening were then downloaded and furthered refined by the authors based on criterion: providing the necessary descriptive information including year of publication, type of blockchain and platform used, and whether a smart contract was used. A total of 41 publications were finally collected for analysis. The number of hits seems somewhat small in light of broad promotion of blockchain in various industries. However, the selected papers, together with the 214 papers searched, forms an instrumental information foundation from which useful findings can be derived. Based on the literature review, a multi-criteria decision matrix (MCDM) was developed through DMM: a method enabling decision makers to identify various factors and analyze the relationships between them (Salmeron & Smarandache, 2008). Firstly, the collected criteria were divided into three categories: type of blockchain-related (Organization criteria); platform-related (Performance criteria); and system-related (Application criteria). Then, categorized criteria further refined through consensus in focus group meetings. The refinement process was repeated to fine-tune the MCDM because some criteria were feasible but not the most advantageous. Besides, Organization criteria were linked to project characteristics to ensure blockchain suitability from the beginning. The resulting MCDM was proposed and graphed. The resulting MCDM is based on the literature and examples in non-project settings. Ideally it should be substantiated by real-life PM practices, but these proved difficult to find as blockchain is quite new in projects. Therefore, we designed a semi-hypothetical study contextualized in the construction industry where projects have been adopted as the norm across a wide range of activities. By “semi-hypothetical”, we mean that our cases are generalized from recurring construction industry phenomena. The scenario is to use 8 blockchain to improve the document endorsement process, one of the most onerous processes in contemporary construction PM, and to use smart contracts to streamline the process automatically and logically. 215 220 225 The case study is divided into three stages. Table 2 lists these stages and the aim, substantiation method and expected outcome of each. Through a 3-part questionnaire and an illustrative example, three criteria dimensions of the MCDM, namely Organization, Performance, and Application were substantiated. Part 0 of the questionnaire gathered the demographic characteristics of the respondents and their firms. In Stage I/Part I of the questionnaire, assuming a construction project client planned to build a blockchain infrastructure. The overarching aim is to use blockchain to enhance the security and traceability of the document endorsement process. The blockchain solution should maintain scalability for different applications (client may use it for multiple projects) and protect user privacy. Thus, respondents were asked questions pertinent to MCDM Organization criteria. It was assumed that respondents familiar with document endorsement in a project setting would have sufficient knowledge to help prospective users select Organization criteria to determine choice of blockchain type. In Stage II/Part II of the questionnaire, questions related to the Performance criteria of the MCDM were asked to substantiate that these criteria could help clients understand which blockchain platform is preferred by project teams. 230 235 We coded the questions in an online survey platform called “Questionnaire Star” (https://www.wjx.cn). A pilot questionnaire was conducted before the full-scale questionnaire. In this study, the target population was clients, contractors, designers, consultants, and researchers serving China’s construction industry. Drawing on the authors’ contacts, the questionnaires were distributed in early August 2020 and recipients invited to respond by mid-August. Stage Start I Aim To gather demographic characteristics of the respondents and their firms To substantiate the Organization criteria of the MCDM Table 2. Semi-hypothetical case study stages Substantiation method Questionnaire (Part 0): 8 multiple-choice questions Expected outcome The information is collected Questionnaire (Part I): each Organization criterion in the MCDM constitutes a multiple-choice question (8 in total). The public, private and consortium blockchain information in Figure 3 corresponding to each Organization criterion constitutes the three options for each question. Respondents were asked to select the most suitable option for each question. The type of blockchain that “wins” is determined by summing the selected options. In this study, it is assumed that the weighting of each Organization criterion is equal. The type of blockchain with the most selected options is adopted 9 240 245 250 II To substantiate the Performance criteria of the MCDM Questionnaire (Part II): each type of blockchain has three alternative platforms in Figure 4 (Hyperledger Fabric is used as both a private and a consortium blockchain platform). Each Performance criterion in MCDM constitutes a multiple-choice question (6 in total), and platform information under each Performance criterion in Figure 4 constitutes options. Respondents evaluated the platforms under the selected type of blockchain by choosing the most suitable option for each question. The final platform is chosen by adding preferences. In this study, it is considered that the weighting of each Performance criterion is equal. The blockchain platform with the most selected options is adopted III To substantiate the Application criteria of the MCDM Completing the tasks corresponding to each Application criteria with information collected from commercial publications A blockchain prototype system is produced Of the questionnaires distributed, we received 110 responses, but some were abandoned due to incompleteness. Finally, 85 were output, edited, and analyzed in Excel spreadsheets. Regarding demographic characteristics, 85% of respondents held a bachelor’s degree or above. About 54% had 1–5 years’ experience in the construction industry, 29% had 6-10 years’ experience, and 17% had more than 10 years’ experience. Companies were a mix of clients, designers, consultants, and main contractors. For Stage III, data collected from commercial publications was used to answer the task questions corresponding to each application criterion, the purpose being to substantiate that the user requirements of a blockchain system can be specified using the Application criteria. Review, proposition, and substantiation Blockchain applications in non-project settings 255 Based on our literature review, Table 3 shows the results of blockchain selection in 41 cases in 10 industrial settings. Among them, 10, 6, and 19 cases adopted public, private, and consortium blockchains, respectively, and 6 adopted a combination. Ethereum was the most popular public blockchain platform. Hyperledger Fabric was the favored consortium blockchain platform and was implemented as a private blockchain in 3 cases. Some scholars built their own new platforms instead of using existing ones. Smart contracts and different consensus mechanisms were widely used. 260 10 ID Reference Table 3. Blockchain choices in the included studies Type of blockchain Platform/ Framework Smart contract Consensus New New Yes N/A Customized PoA Ethereum New Hyperledger Fabric Ethereum Yes Yes Yes Yes PoW BFT BFT PoW Hyperledger Fabric Hyperledger Fabric New Yes Yes Yes N/A PBFT PoW Hyperledger Fabric Ethereum New Hyperledger Fabric Yes Yes Yes Yes PBFT N/A Customized PBFT JUICE New New New N/A Yes Yes N/A PoC PBFT DPoS PoW/PBFT New Hyperledger Fabric New New New Yes Yes Yes Yes Yes PoS Customized PBFT PBFT MCPBFT Quorum Yes v v v Hyperledger Fabric Hyperledger Fabric New Yes Yes Yes v New New New N/A Yes Yes Quorum Consensus Kafka Consensus BFT AlgoRand consensus PoAh PoA Customized Ark Hyperledger Fabric Hyperledger Fabric Hyperledger Fabric; Ethereum N/A Yes N/A Yes DPOS N/A N/A PoW/PBFT/ BFT Ethereum New Ethereum New Ethereum Yes Yes Yes Yes Yes N/A PoI PoS ADMM N/A Bitcoin Hyperledger Fabric New N/A N/A Yes PoW OMMS PBFT Public PrivateConsortium Intellectual property and digital identity 1 Kim and Yu (2018) v 2 Halloush and Yaseen v v (2019) Philanthropy, aid, and donors 3 Nor et al. (2017) v 4 Sirisha et al. (2019) v 5 Lamba et al. (2019) v 6 Saleh et al. (2019a) v Democracy and governance 7 Zhang et al. (2018) v 8 Yu et al. (2018) v 9 Diallo et al. (2018) v Agriculture 10 Ge et al. (2017) v 11 Shih et al. (2019) v 12 Yang et al. (2020) v 13 Hang et al. (2020) v Health 14 Zhang and Lin (2018) v v 15 Griggs et al. (2018) v 16 Liu et al. (2018) v 17 Xu et al. (2019) v v Logistics and supply chain management 18 Kaijun et al. (2018) v 19 Ding et al. (2019) v v 20 Mao et al. (2019) v 21 Chen et al. (2020) v 22 Li et al. (2020) v Built environment 23 Brousmiche et al. (2018) v 24 25 26 Mukne et al. (2019) She et al. (2019) Zhang and Wang (2019) 27 Puthal et al. (2020) 28 Kaur and Oza (2020) 29 Luo et al. (2019) Education and human rights 30 Turkanović et al. (2018) 31 Choudhury et al. (2018) 32 Saleh et al. (2019b) 33 Wang et al. (2019) Environment and energy 34 Niya et al. (2018) 35 Mengelkamp et al. (2018) 36 Pop et al. (2018) 37 Van Leeuwen et al. (2020) 38 França et al. (2020) Finance and economy 39 Chen et al. (2017) 40 Wang et al. (2018a) 41 Xu and Viriyasitavat (2019) v v v v v v v v v v v v v v v v v 11 Note: New=building a custom blockchain or forking an established blockchain; BFT= Byzantine fault tolerant; MCPBFT=Multi-center practical byzantine fault tolerance; ADMM=Alternating direction method of multipliers; OMMS=Orchestration of multiple modules and steps; PoI= Proof of identity; N/A= Not available 265 270 A proposed multi-criteria decision matrix Figure 2 shows our MCDM. Organization criteria help interested users choose a blockchain type based on project characteristics; Performance criteria help evaluate existing blockchain platforms; and Application criteria assist in planning blockchain systems. Users should follow the order from Organization, Performance to Application criteria to use MCDM. Figure 2. The MCDM for choosing blockchain for projects 275 280 Figure 3 shows public, private, and consortium blockchain information corresponding to Organization criteria. The first step is to define the range of blockchain use, project transparency, and centralization levels. The next is to determine whether a project should be highly auditable or reserve some space to retain sensitive business information. After that, we estimate the cost of a blockchain solution. For example, according to Ernst & Young (2019), private blockchains are usually adopted in projects with high transaction volumes because high costs are averaged over a large number of transactions. Finally, we determine the privacy, security and scalability requirements in projects. The notes in Figure 3 elaborate upon the Organization criteria. 12 285 Figure 3. Organization criteria for choosing blockchain types 290 295 Figure 4 elaborates on the MCDM Performance criteria applied to 8 popular blockchain platforms. The first criterion, transaction throughput, refers to the number of transactions the blockchain can commit per second. Block time is the time required to generate the next block in a chain. Regarding consensus network, depending on the use case, a platform must support single or multiple consensuses, or a pluggable consensus. The number of languages supporting smart contract writing and the need for cryptocurrency should then be evaluated. Lastly, the cost per transaction should be considered. 13 Figure 4. Performance criteria for choosing blockchain platforms 300 305 Figure 5 elaborates on the Application criteria used to specify user requirements based on use cases so that developers can better plan blockchain systems. Blockchain system architecture normally consists of six layers: application, blockchain as a service (BaaS), execution, consensus, data, and network. The consensus layer managing the agreement between stakeholders through various consensus algorithms and the data layer, which handles the blockchain content organization using components such as hash functions, chain structure, and timestamps, are hosted by existing or newly built platforms. So for successful system development, one should perform tasks corresponding to the Application criteria to specify user requirements for the other four layers. 14 310 Figure 5. Application criteria for planning blockchain systems 315 320 Using the Project structure, we can first specify what applications in the application layer will be implemented with blockchain and determine the number of nodes and organizations that will participate in the network layer. The Transaction definition aims to determine specific transactions so that applications can interact with the blockchain network by submitting transactions to the ledger or querying the ledger’s contents. Based on the defined transaction and the mutual trust relationship of nodes, the Communication channel criterion helps plan the number of communication channels (if dedicated channels are available). In this way, the BaaS layer, especially its various software development kits (SDKs), can realize the management, access control, and communication between the network and application layers. Finally, the Smart contract function helps specify its scope and the functions it needs to perform at the execution layer. 15 325 Substantiation via a semi-hypothetical case study Stage I—Organization criteria 330 335 Figure 6 shows the results of evaluating blockchain types according to the Organization criteria. Of the questionnaire respondents, 55% chose a consortium blockchain between multiple authorized organizations in a construction project; 65% preferred information to be transparent among all involved organization members or certain sub-members; and 61% believed that partial centralization in the consortium blockchain could meet their requirements. For auditability, 60% chose the auditability of the public blockchain, and 40% accepted that its “cost and transaction volume are positively correlated”. Regarding project privacy, 50% chose the private blockchain’s privacy. For security, 60% chose the security of the public blockchain. Regarding scalability, 41% believe that the scalability of private blockchains can meet the needs of construction projects, while 43% believe that moderate scalability of consortium blockchains is appropriate. Only 16% preferred the scalability of public blockchains. 340 Figure 6. Evaluation results of blockchain type against the Organization criteria 345 Answering the 8 questions on Organization criteria, 55%, 25%, and 18% respondents chose more options corresponding to the consortium, public and private blockchain, respectively (1% had no preference; 1% chose both public and consortium blockchains). Based on this result, consortium blockchain is most feasible for this study. 16 Stage II—Performance criteria 350 355 360 In Stage II, according to Stage I results, 56%, 26%, and 18% evaluations based on Performance criteria were received for consortium, public and private blockchain platforms, respectively (1 evaluated all platforms; 1 assessed both public and consortium blockchain platforms). Evaluation results of consortium blockchain platforms R3 Corda, Hyperledger Fabric, and Ark are shown in Figure 7a. For transaction throughput, 45% chose Hyperledger Fabric. The block time preference was R3 Corda (57%). More than 67% selected the pluggable consensus of Hyperledger Fabric, 88% because it supports multiple languages for writing smart contracts Regarding cryptocurrency, 37% did not need the platform to provide it (R3 Corda do not have a native cryptocurrency), 49% thought it might be needed (Hyperledger Fabric can provide development opportunities), and only 14% specified that a platform should have a native cryptocurrency (i.e., Ark). For transaction fee, 43% preferred R3 Corda, 51% chose Hyperledger Fabric, and 6% Ark. Overall, more respondents chose options corresponding to Hyperledger Fabric. 17 365 Figure 7. Evaluation results of blockchain platforms against the Performance criteria 370 375 Evaluation results of public and private blockchain platforms are shown in Figures 7b and 7c, respectively. For public blockchain platforms, more respondents chose Bitcoin’s transaction throughput and block time (48% and 39%, respectively), EOS’s consensus network (40%) and transaction fees (48%), and Ethereum’s smart contract language (48%). For native cryptocurrency, respondents specified that the platform should have it (65% designated Bitcoin, 26% Ether, and 9% EOS). Overall, more respondents chose options corresponding to 18 380 Bitcoin. Those choosing private blockchain leant towards Hyperledger Sawtooth’s transaction throughput (56%) and consensus network (56%), Fabric’s block time (44%), and NEO’s smart contract language (63%). For native cryptocurrency, 31% respondents specified that a platform should have it (e.g., NEO), 69% thought it might not be needed (e.g., Hyperledger Sawtooth), and none answered “it might be needed” (e.g., Hyperledger Fabric). Since none of these three private platforms charge transaction fees, the “transaction fee” criterion was not used. Overall, Hyperledger Sawtooth was more selected. Stage III—Application criteria 385 Based on Stages I and II, Stage III collects user requirements of the Hyperledger Fabric system by deploying Application criteria. By referring to commercial publications and scenario building in the construction industry, an example of how to perform tasks related to each application criterion is given (Table 4). In this way, developers can plan a user-centered system. 390 Table 4. User requirements in a typical construction project 395 Application criteria Tasks Project structure Applications: Document endorsement, Smart contracting Blockchain architecture Application Four organizations: Client, designer, consultant, and contractor Four peer nodes: Project leader, Architect, inspector, project manager Two ordering nodes: Administration points (assigned by both client and contractor) Network Transaction definition Preliminary design plan, Final design submission, Design and construction changes, modifications, worked hours, additional work claims, Quality assurance approvals, Key performance indicator (KPI) inquiring Application Communication channel Two channels: Channel C1 for preliminary design plan related documents, and Channel C2 for other transactions BaaS Smart contract function Scope: Smart contract S1 executes business logic in C1 and smart contract S2 defines executable logic in C2 Execution Functions: Collect approvals, Check, Report and update KPI, Notify results Execution Figure 8 shows the planned blockchain system. In this study, Client, Designer, Consultant, and Contractor agree that they will use and build the consortium blockchain in Hyperledger Fabric for document endorsement and smart contracting. The Client is the network initiator. Organizations can have applications that can perform transactions in two channels. The Client 19 400 405 and Designer need to communicate about the design plan privately via an isolated channel. Peer nodes Project leader and Architect maintain a copy of the Ledger L1 associated with Channel C1 for the preliminary design plan and maintain a copy of the Ledger L2 associated with Channel C2 for other transactions indicated in Table 4. The peer nodes Project manager and Inspector maintain a copy of the Ledger L2 associated with Channel C2. There are two Ordering services O1 and O2 that service as network administration points for network and support application Channels C1 and C2. Smart contract S1 is installed on Project leader and Architect to execute the logic of business in C1. Smart contract S2 is installed on all peer nodes to define the executable logic for C2. Figure 8. Final system configuration against the Application criteria 410 415 The network is managed according to rules specified in Network Configuration NC4 (e.g., both Client and Contractor need to agree if new organizations want to join the network). When it comes to consensus, transactions in Channel C1 are managed according to rules specified in channel configuration CC1 (only the Project leader and Architect need to endorse the preliminary design plan-related documents); the Client and Designer manage C1. All organizations manage Channel C2 according to Rules CC2 (all peer nodes need to endorse transactions). Organizations have certificate authorities defined in the data layer. This successful planning of a blockchain system substantiates our Application criteria feasibility. Discussion 420 The MCDM developed in this study provides a structured methodology to help prospective PM professionals choose suitable blockchain and plan implementable blockchain-based PM systems. Organization, Performance, and Application pertinent criteria have been investigated 20 425 430 435 440 in various industries or non-project settings, but this study links the criteria with projects and enriches their meaning in PM discourse. Our case study provides a vivid example of how the MCDM can be implemented to choose a blockchain and plan a blockchain system for a client in construction projects. Respondents in this study had different blockchain projects in mind when they completed the surveys. In real-life practice, project clients may wish to build a blockchain infrastructure for different applications. Therefore, the proposed MCDM can help project clients mine the actual requirements of different project teams and then decide on blockchain options. Overall, this research is a meaningful step towards demystifying blockchain in the project world. Blockchain presents enormous prospects and challenges for project management. At a technical level, the hash algorithms, distributed ledgers, consensus mechanisms, and other components are still developing rapidly. At the service level, smart contracts are probably the most promising application for safeguarding information, expediting reporting and endorsement, and paying on time. The hash algorithm, distributed database, and decentralized consensus mechanism make it difficult for any party in the blockchain to tamper with the project’s information. However, one cannot underestimate security vulnerability or make assumptions about blockchain risks, such as 50% vulnerabilities, code vulnerabilities, private key security, criminal activity, and identity exposure (Perera et al., 2020). A typical misunderstanding is that the technology can guarantee the genuineness of information in the off-chain, project world. Traditional quality assurance and inspection technologies in the physical project world are still needed to enable blockchain power. 445 450 455 460 This research has some limitations. Firstly, the weighting of each Organization and Performance criterion were assumed to be equally important. Secondly, in the case study, criteria (Organization, Performance, and Application) were adopted independently. Thirdly, in this research, the proposed MCDM was used as a decision-support tool to help users choose blockchain options, thereby not naturally considering detailed information governance of projects. Fourthly, PM professionals would ideally evaluate systems based on real-life construction projects, but such projects using blockchain are few and far between. We developed a prototype of the system on the Hyperledger Fabric platform. But in the case study, the blockchain system, like any blockchain system in various industries, mainly works on the background. Evaluators do not even notice such system being at play. The unique terminologies further prohibited the users from evaluating the system sensibly. Amid the global blockchain hype, a word of caution is that prospective users should carefully choose the parts of their PM tasks to be blockchained, and to ask whether they really need blockchain at all. The technology introduces redundancy (e.g., by placing the blocks in 21 465 multiple places) and sacrifices efficiency for improved security (Xue & Lu, 2020). It incurs extra, sometimes excessive, cost. Therefore, a highly selective strategy is desired based on more thorough cost-benefit analyses when better empirical data is available. Blockchain implementation is not just about using new software. It is more about implementing novel business technologies and philosophies (Penzes et al., 2018). It has the potential to support digitization in projects and provide solutions to many challenges, but before using the technology, organizations need to dissect their existing business models. Therefore, we need to understand more about the in-depth relationship between project governance and blockchain system and their impacts on project performance. 470 Conclusions 475 480 485 490 495 Project management (PM) professionals are exploring the potential of blockchain, a disruptive technology promising benefits such as greater transparency, enhanced security, improved tractability, and increased efficiency. However, no previous study examines blockchain technologies suitable for projects, whose unique characteristics may impose particular challenges to this emerging technology. This research addresses this knowledge gap by proposing a multi-criteria decision matrix (MCDM) for PM professionals to choose blockchain for their projects. The MCDM has three staged components organized sequentially. Stage I is to choose a generic blockchain type (e.g., public, private, or consortium) against organization-related criteria range of use, transparency, centralization, auditability, cost, privacy, security, and scalability. Stage II is to choose a specific blockchain platform against performance-related criteria transaction throughput, block time, consensus network, smart contract language, cryptocurrency, and transaction fee. Stage III is to specify user requirements to plan a blockchain system against the application-related criteria project structure, transaction definition, communication channel, and smart contract function. The MCDM was tested and evaluated through a case study that includes a questionnaire and an illustrative example pertinent to the construction industry. The results show that consortium blockchain is superior to public or private blockchains in dealing with the unique characteristics of projects. Hyperledger Fabric, a commercial platform, was chosen as the best solution. Based on the platform, developers need to consider user requirements such as network participants, principal transactions, communication channels, and smart contracts to realize a blockchain-based PM system. This is one of the first studies to truly “soft land” the technology in projects as a widespread yet unique organization and governance setting. It makes several original contributions. Firstly, it examined the challenges arising from project characteristics when choosing a type of blockchain. Secondly, it isolated blockchain selections in 10 industrial settings and 22 500 505 510 515 520 identified the criteria for selecting blockchain types/platforms. Thirdly, it guides system developers to consider both the blockchain type/platform and the project characteristics/processes in developing a blockchain system. Not only useful for demystification purposes, the MCDM is a readily deployable methodology for developing blockchain systems for PM. The limitations of this study provide opportunities for further investigations. Firstly, the weighting of each Organization and Performance criterion were assumed to be equal. More research is desired to assess the relative importance of the criteria by considering more reallife projects. Future research can also link criteria and their weightings to business outcomes. Therefore, users can understand the value that blockchain can instigate before implementation. Secondly, criteria (Organization, Performance, and Application) were adopted independently in the case study. With the rapid development of blockchains, understanding the interdependence of the criteria can make the MCDM more robust in the face of new capabilities. Thirdly, the proposed MCDM is an infrastructure for selecting blockchain options, thereby not naturally considering complicated information governance. Therefore, future investigations are encouraged to form logical information structures so that project participants can determine how important issues such as data privacy level and transaction throughput exactly are for their blockchain use cases. Lastly, in-depth investigations of the relationship between project governance and blockchain system and their impacts on project performance are much desired when more cases are available. Blockchain is novel, and there are numerous challenges, but the potential of this technology to reshape the PM domain is too great to be missed. Acknowledgements 525 The work presented in this paper was financially supported by the Hong Kong Innovation and Technology Commission (ITC) with the Innovation and Technology Fund (ITF) (No. ITP/029/20LP). Data availability statement 530 The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to their containing information that could compromise the privacy of research participants. 23 535 540 545 550 555 560 565 570 575 580 References Allen, T. J. (1984). Managing the flow of technology: Technology transfer and the dissemination of technological information within the R&D organization. Boston: The MIT Press. Ark. (2019). ARK Whitepaper. Retrieved April 18, 2020, from ARK: https://ark.io/whitepaper Allen, D. W., Berg, C., & Novak, M. (2018). Blockchain: an entangled political economy approach. Journal of Public Finance and Public Choice, 33(2), 105-125. Brousmiche, K. L., Heno, T., Poulain, C., Dalmieres, A., & Hamida, E. B. (2018). Digitizing, securing and sharing vehicles life-cycle over a consortium blockchain: Lessons learned. 2018 9th IFIP International Conference on New Technologies, Mobility and Security (pp. 1-5). IEEE. Chen, J., Cai, T., He, W., Chen, L., Zhao, G., Zou, W., & Guo, L. (2020). A blockchain-driven supply chain finance application for auto retail industry. Entropy, 22(1), 95. Choudhury, O., Sarker, H., Rudolph, N., Foreman, M., Fay, N., Dhuliawala, M., ... & Das, A. K. (2018). Enforcing human subject regulations using blockchain and smart contracts. Blockchain in Healthcare Today, 1, 1-14. Chen, P. W., Jiang, B. S., & Wang, C. H. (2017). Blockchain-based payment collection supervision system using pervasive Bitcoin digital wallet. 2017 IEEE 13th International Conference on Wireless and Mobile Computing, Networking and Communications (pp. 139-146). IEEE. Diallo, N., Shi, W., Xu, L., Gao, Z., Chen, L., Lu, Y., ... & Turner, G. (2018). eGov-DAO: A better government using blockchain based decentralized autonomous organization. 2018 International Conference on eDemocracy & eGovernment (pp. 166-171). IEEE. Ding, Q., Gao, S., Zhu, J., & Yuan, C. (2019). Permissioned blockchain-based double-layer framework for product traceability system. IEEE Access, 8, 6209-6225. DeFillippi, R., and Sydow, J. (2016). Project networks: Governance choices and paradoxical tensions. Project Management Journal, 47(5), 6-17. Davidson, S., De Filippi, P., & Potts, J. (2018). Blockchains and the economic institutions of capitalism. Journal of Institutional Economics, 14(4), 639-658. Ernst & Young. (2019). Total cost of ownership for blockchain solutions. Ernst & Young LLP. Engwall, M. (2003). No project is an island: Linking projects to history and context. Research Policy, 32(5), 789–808. França, A. S. L., Neto, J. A., Gonçalves, R. F., & Almeida, C. M. V. B. (2020). Proposing the use of blockchain to improve the solid waste management in small municipalities. Journal of Cleaner Production, 244, 118529. Ge, L., Brewster, C., Spek, J., Smeenk, A., Top, J., van Diepen, F., ... & de Wildt, M. D. R. (2017). Blockchain for agriculture and food: Findings from the pilot study (No. 2017-112). Wageningen Economic Research. Griggs, K. N., Ossipova, O., Kohlios, C. P., Baccarini, A. N., Howson, E. A., & Hayajneh, T. (2018). Healthcare blockchain system using smart contracts for secure automated remote patient monitoring. Journal of Medical Systems, 42(7), 130. Hyperledger. (2020). Glossary. Retrieved April 15, 2020, from Hyperledger Fabric: https://hyperledger-fabric.readthedocs.io/en/latest/index.htm. Halloush, Z. A., & Yaseen, Q. M. (2019). A blockchain model for preserving intellectual property. Second International Conference on Data Science, E-Learning and Information Systems (pp. 1-5). Association for Computing Machinery. Hang, L., Ullah, I., & Kim, D. H. (2020). A secure fish farm platform based on blockchain for agriculture data integrity. Computers and Electronics in Agriculture, 170, 105251. 24 585 590 595 600 605 610 615 620 625 ISO. (2020). Blockchain and distributed ledger technologies — Vocabulary. Retrieved May 8, 2021, from Technical Committee ISO/TC 307: https://www.iso.org/obp/ui/#iso:std:iso:22739:ed-1:v1:en. Kim, E. J., & Yu, J. (2018). A consortium blockchain system for verifying digital contents on traditional costumes. Euro-Mediterranean Conference (pp. 437-446). Springer. Kaijun, L., Ya, B., Linbo, J., Han-Chi, F., & Van Nieuwenhuyse, I. (2018). Research on agricultural supply chain system with double chain architecture based on blockchain technology. Future Generation Computer Systems, 86, 641-649. Kaur, E., & Oza, A. (2020). Blockchain-based multi-organization taxonomy for smart cities. SN Applied Sciences, 2(3), 1-11. Lu, W., Xu, J., & Söderlund, J. (2020). Exploring the effects of building information modeling on projects: Longitudinal social network analysis. Journal of Construction Engineering and Management, 146(5), 04020037. LeewayHertz. (2019). Top blockchain platforms to watch out for in 2019. Retrieved April 18, 2020, from Hacker Noon: https://hackernoon.com/top-blockchain-platforms-to-watch-outin-2019-aa80e336a426 Lamba, R., Gupta, Y., Kalra, S., & Sharma, M. (2019). Preventing waiting list manipulation and black marketing of donated organs through Hyperledger Fabric. 2019 International Conference on Computing, Communication, and Intelligent Systems (ICCCIS) (pp. 280285). IEEE. Li, X., Lv, F., Xiang, F., Sun, Z., & Sun, Z. (2020). Research on key technologies of logistics information traceability model based on consortium chain. IEEE Access, 8, 69754-69762. Liu, J., Li, X., Ye, L., Zhang, H., Du, X., & Guizani, M. (2018). BPDS: A blockchain based privacy-preserving data sharing for electronic medical records. 2018 IEEE Global Communications Conference (pp. 1-6). IEEE. Luo, H., Das, M., Wang, J., & Cheng, J. C. P. (2019). Construction payment automation through smart contract-based blockchain framework. International Symposium on Automation and Robotics in Construction (pp. 1254-1260). IAARC Publications. Lee, J. Y. (2019). A decentralized token economy: How blockchain and cryptocurrency can revolutionize business. Business Horizons, 62(6), 773-784. Li, X., Wu, L., Zhao, R., Lu, W., & Xue, F. (2021). Two-layer Adaptive Blockchain-based Supervision model for off-site modular housing production. Computers in Industry, 128, 103437. Mao, D., Hao, Z., Wang, F., & Li, H. (2019). Novel automatic food trading system using consortium blockchain. Arabian Journal for Science and Engineering, 44(4), 3439-3455. Mukne, H., Pai, P., Raut, S., & Ambawade, D. (2019). Land record management using Hyperledger Fabric and IPFS. 2019 10th International Conference on Computing, Communication and Networking Technologies (pp. 1-8). IEEE. Mengelkamp, E., Notheisen, B., Beer, C., Dauer, D., & Weinhardt, C. (2018). A blockchainbased smart grid: Towards sustainable local energy markets. Computer Science-Research and Development, 33(1-2), 207-214. Nor, R. M., Rahman, M. H., Rahman, T., & Abdullah, A. (2017). Blockchain sadaqa mechanism for disaster aid crowd funding. The 6th International Conference on Computing and Informatics: Embracing Eco-Friendly Computing (pp. 25-27). Universiti Utara Malaysia. Niya, S. R., Jha, S. S., Bocek, T., & Stiller, B. (2018). Design and implementation of an automated and decentralized pollution monitoring system with blockchains, smart contracts, and LoRaWAN. NOMS 2018-2018 IEEE/IFIP Network Operations and Management Symposium (pp. 1-4). IEEE. 25 630 635 640 645 650 655 660 665 670 675 Penzes, B., Kirkup, A., Gage, C., Dravai, T., & Colmer, M. (2018). Blockchain technology in the construction industry: Digital transformation for high productivity. Institution of Civil Engineers. Perera, S., Nanayakkara, S., Rodrigo, M. N. N., Senaratne, S., & Weinand, R. (2020). Blockchain technology: Is it hype or real in the construction industry? Journal of Industrial Information Integration, 100125. Powell, W. W. (1990). Neither market nor hierarchy. In M. Godwyn, & J. H. Gittell, Sociology of Organizations: Structures and Relationships (pp. 104-117). SAGE Publications. Project Management Institute. (2000). A guide to the project management body of knowledge. 4th ed. Project Management Institute. Puthal, D., Mohanty, S. P., Yanambaka, V. P., & Kougianos, E. (2020). PoAh: A novel consensus algorithm for fast scalable private blockchain for large-scale IoT frameworks. arXiv:2001.07297. Pop, C., Cioara, T., Antal, M., Anghel, I., Salomie, I., & Bertoncini, M. (2018). Blockchain based decentralized management of demand response programs in smart energy grids. Sensors, 18(1), 162. Salmeron, J. L., & Smarandache, F. (2008). Redesigning decision matrix method with an indeterminacy-based inference process. International Journal of Applied Mathematics & Statistics, 13(8), 4-11. Sirisha, N. S., Agarwal, T., Monde, R., Yadav, R., & Hande, R. (2019). Proposed solution for trackable donations using blockchain. 2019 International Conference on Nascent Technologies in Engineering (pp. 1-5). IEEE. Saleh, H., Avdoshin, S., & Dzhonov, A. (2019a). Platform for tracking donations of charitable foundations based on blockchain technology. 2019 Actual Problems of Systems and Software Engineering (APSSE) (pp. 182-187). IEEE. Shih, D. H., Lu, K. C., Shih, Y. T., & Shih, P. Y. (2019). A simulated organic vegetable production and marketing environment by using Ethereum. Electronics, 8(11), 1341. She, W. E. I., Gu, Z. H., Lyu, X. K., Liu, Q. I., Tian, Z., & Liu, W. (2019). Homomorphic consortium blockchain for smart home system sensitive data privacy preserving. IEEE Access, 7, 62058-62070. Saleh, O. S., Ghazali, O., & RANA, M. E. (2019b). Blockchain based framework for educational certificates verification. Journal of Critical Reviews, 7(3), 2020. Shilov, K. (2018). When is it time to build your own blockchain? Retrieved May 30, 2020, from Hacker Noon: https://medium.com/hackernoon/when-is-it-time-to-build-your-ownblockchain-f3be0a30b826 Smyth, H. (2018). Projects as creators of the preconditions for standardized and routinized operations in use. International Journal of Project Management, 36(8), 1082-1095. Sydow, J., Schüßler, E., and Müller-Seitz, G. (2016). Managing inter-organizational relations: Debates and cases. London: Palgrave Macmillan. Turkanović, M., Hölbl, M., Košič, K., Heričko, M., & Kamišalić, A. (2018). EduCTX: A blockchain-based higher education credit platform. IEEE Access, 6, 5112-5127. Turner, J. R., & Müller, R. (2003). On the nature of the project as a temporary organization. International Journal of Project Management, 21(1), 1-8. Taghizadeh-Hesary, F., & Yoshino, N. (2019). The way to induce private participation in green finance and investment. Finance Research Letters, 31, 98-103. Van Leeuwen, G., AlSkaif, T., Gibescu, M., & Van Sark, W. (2020). An integrated blockchainbased energy management platform with bilateral trading for microgrid communities. Applied Energy, 263, 114613. Winch, G. (1989). The construction firm and the construction project: A transaction cost approach. Construction Management and Economics, 7(4), 331-345. 26 680 685 690 695 700 705 710 Wright, J. A., & Kersner, M. (2004). Short‐term projects: The standards fund and collaboration between speech and language therapists and teachers. Support for Learning, 19(1), 19-23. Wang, G., Zhang, H., Xiao, B., Chung, Y. C., & Cai, W. (2019). EduBloud: A blockchainbased education cloud. 2019 Computing, Communications and IoT Applications (pp. 352357). IEEE. Wang, X., Xu, X., Feagan, L., Huang, S., Jiao, L., & Zhao, W. (2018a). Inter-bank payment system on enterprise blockchain platform. 2018 IEEE 11th International Conference on Cloud Computing (pp. 614-621). IEEE. Wang, H., Lu, W., Söderlund, J., and Chen, K. (2018b). The interplay between formal and informal institutions in projects: A social network analysis. Project Management Journal, 49(4), 20-35. Xu, J., Xue, K., Li, S., Tian, H., Hong, J., Hong, P., & Yu, N. (2019). Healthchain: A blockchain-based privacy preserving scheme for large-scale health data. IEEE Internet of Things Journal, 6(5), 8770-8781. Xu, L., & Viriyasitavat, W. (2019). Application of blockchain in collaborative Internet-ofThings services. IEEE Transactions on Computational Social Systems, 6(6), 1295-1305. Xue, F., & Lu, W. (2020). A semantic differential transaction approach to minimizing information redundancy for BIM and blockchain integration. Automation in Construction, 118, 103270. Yu, B., Liu, J. K., Sakzad, A., Nepal, S., Steinfeld, R., Rimba, P., & Au, M. H. (2018). Platformindependent secure blockchain-based voting system. International Conference on Information Security (pp. 369-386). Springer. Yang, H., Xiong, S., Frimpong, S. A., & Zhang, M. (2020). A consortium blockchain-based agricultural machinery scheduling system. Sensors, 20(9), 2643. Zhang, W., Yuan, Y., Hu, Y., Huang, S., Cao, S., Chopra, A., & Huang, S. (2018). A privacypreserving voting protocol on blockchain. 2018 IEEE 11th International Conference on Cloud Computing (pp. 401-408). IEEE. Zhang, A., & Lin, X. (2018). Towards secure and privacy-preserving data sharing in e-health systems via consortium blockchain. Journal of Medical Systems, 42(8), 140. Zhang, X., & Wang, D. (2019). Adaptive traffic signal control mechanism for intelligent transportation based on a consortium blockchain. IEEE Access, 7, 97281-97295. 27