A forest-fire model for technology risk transfer

Abstract

Based on the transmission characteristics of networks and the systematization of technical risks in national major science and technology projects, the risk transmission mechanism of national major science and technology projects is deeply discussed herein. Firstly, the system of systems (SoS) engineering process model of national major science and technology projects is constructed, including two levels and three stages. Secondly, the hierarchical structure of national major science and technology projects is analyzed, and the risks are divided into different levels, such as SoS, system, subsystem, equipment, module, and components. Finally, we describe the risk transfer mechanism using the forest-fire model and outline a risk control strategy. The results show that the findings are helpful for recognizing the essence and transmission mechanism of technical risks of national major science and technology projects and for reducing the risk from the source. The research results can be applied to project management in the transportation field.

Keywords

SoS engineering risk transfer forest-fire model

1 Introduction

1.1 Major scientific and technological projects

Major scientific and technological projects are major strategic products, key common technologies, and major projects completed within a certain time limit through core technology breakthroughs and resource integration to achieve national goals, such as the two bombs and one satellite project, Beidou satellite navigation system, manned aerospace and lunar exploration projects, high-resolution earth observation systems, large aircraft, etc. In November 2015, the Recommendations of the Central Committee of the Communist Party of China on Formulating the Thirteenth Five-Year Plan for National Economic and Social Development state that, facing 2030, a number of major scientific and technological projects and major projects that reflect the national strategic intentions will be implemented to strive to achieve breakthroughs in communication and quantum computing, integrated space-to-earth information networks, intelligent manufacturing and robotics, artificial intelligence 2.0, and other key directions. It forms a system layout with the major national science and technology projects implemented as of 2006 (involving electronic information, advanced manufacturing, energy and environmental protection, biomedicine, etc.).

The construction of major science and technology projects is susceptible to changes in demand, development goals, the external environment, departmental communication and coordination, and other factors, posing new challenges and uncontrollable risks for engineering construction, and resulting in delays in construction schedules, cost overruns, and poor communication and coordination. The challenges and uncontrollable risks of the project can increase the complexity of the project’s management objectives, the management environment, management objectives, organizational structure, organizational behaviors, etc. It is difficult for managers to dynamically monitor the project development process from all aspects.

1.2 Related research

Current research on the risks of major science and technology projects is mainly focused on extending the theory and technology of risk management to major science and technology projects, including risk planning, identification, evaluation, response, and monitoring. Lee et al. [8] proposed using a Bayesian network to manage risks in the construction of large engineering projects. Grimsey et al. [7] analyzed the risks of PPP(Public Private Partnership) projects from the perspectives of different stakeholders by analyzing the Scottish AV&S (Almond Valley and Seafield) project, and summarized the risk evaluation methods, evaluation perspectives, evaluation indicators, and the risks faced by PPP projects. Wibowo et al. [2] used a cube stochastic model to simulate and systematically analyze the risks of PPP projects from the perspective of each project participant. Thomas et al. [3] analyzed the impact of risks on the project and the probability of risk on the basis of the Delphi and fault tree methods, and the research framework was mainly divided into two parts: creating a mathematics model based on the main risks in the project, and using expert scores to establish a matrix for judging risks. Research on risk transmission mainly focuses on financial risk and crisis. Taimur et al. [6] studied the risk transmission of countries during the Southeast Asian financial crisis. Kaminsky et al. [10] studied the transmission of credit risk between different countries. Anjali et al. examined [1] the risk profiles of projects based on project characteristics and provided key managerial insights. Yaniv [14] formulated the risk optimization model, which sought to minimize risk through the selection of risk handling strategies. Project Management Strategies for Complex Projects (R10) [14, 18] provides practical tools and techniques to optimize innovation, minimize schedule and budget risks, and build better projects. It expands the three-dimensional analysis typically used by departments of transportation and creates a model that facilitates project management in five areas—cost, schedule, technical, financial, and context; using a step-by-step approach that includes: defining critical project success factors, assembling a project team, selecting project arrangements, preparing early cost model and finance plan, and developing project action plans.

Domestic scholars have conducted large amounts of research on risk analysis. Cui et al. [16] studied the transmission mechanism between micro risks in an enterprise and macro risks during a financial crisis through the analysis of the Asian financial crisis. Zhu et al. [21] demonstrated three main transmission mechanisms of financial risk. Other scholars [9 , 15] mainly studied the definition, path, carrier, and the characteristics and mechanism of risk transmission. Leng et al. [20] focused on the risks of time, technology, member change, coordination, and target change, so that professional project managers can participate in project risk management as soon as possible. Chen et al. [23] adopted the multidimensional balanced scorecard approach to improve project management and organizational performance, to help reduce project risks and potential disasters to maximize the investment benefits and security high-risk IT projects. Liu et al. [5] divided risks into government risk, market risk, and project risk, and the improved matter element model was established to provide effective support for effective project risk management, contract management, and other work. Ren et al. [24] constructed a four-dimensional model of risk management suitable for the characteristics of major scientific research projects. Tang et al. [22] researched the risk transfer model of real estate projects. Shao et al. [17] identified risks existing in the whole process of the scientific research project, and built the risk evaluation model based on the analytic hierarchy process (AHP).

1.3 Problems

The technical risks of major scientific and technological projects have considerably liquidity and transmissibility, which must be comprehensively identified and effectively controlled in the process of engineering construction. The organization of major scientific and technological projects is jointly undertaken by many cross-domain and cross-unit R&D departments; thus, potential problems posed by risks in the major scientific and technological projects include goals constraints and risk factors.

The first problem is goals constraints. The project aims to achieve various goals, including demand, quality, cost, schedule, risk, resources, and information goals at the management level; technology, economy, safety, and usability goals at the functional level; and social, economic, environmental, and industrial goals at the benefit level. Due to the different goals of the overall unit and system undertaking units, the individual and overall goals are prone to conflicts and contradictions driven by their respective interests, which affect and restrict the overall goal.

The second problem is risk factors. In the process of project management, there are many construction process and technology risk factors, and increasing economic and market risks. If those risk factors are not properly controlled, the consequences will be serious.

The development process of major scientific and technological projects covers multiple departments, levels, and stages, making the transmissibility of risks strong. The risks have contagious effects through accumulation, amplification, and even mutation, and eventually lead to a crisis for the whole project. Therefore, to better control project risks, it is necessary to study the risk transmission mechanism of major science and technology projects, and perform early risk prediction and risk control throughout the whole process of the project.

2 System of systems (SoS) engineering process model

Major science and technology projects are characterized by huge engineering scale, complex engineering technology, many participating units, long construction period, and wide spread. The implementation of projects needs to tackle major scientific and technological problems and cutting-edge high-technology, requiring joint research and development from many units. The involved aspects include quality, cost, schedule, safety, organization, information, environment, risk, communication, and so on. The internal and external links and constraints of the project are complex, affecting the political, social,and ecological aspects of the region or country, and having far-reaching impacts on national security, international status, and socio-economic development.

The construction and management of major scientific and technological projects belongs to an open giant system with multiple dimensions, levels, and subsystems, which involves complex system engineering. So, it is necessary to break through traditional system thinking and system engineering methods and use SoS engineering methods to perform the construction and management of major scientific and technological projects.

The SoS engineering process can be generally summarized as two levels and three stages. The two levels refer to the SoS engineering process being divided into two levels: SoS level and system level. The three stages refers to the SoS engineering process mainly including three stages: top-level design, system development, and integration verification, as shown in Fig. 1.

Fig. 1

System engineering process model.

At the SoS level, with system capability generation as the goal, the two stages mainly include top-level design and integration verification. The engineering activities of top-level design are driven by requirements, which mainly involve system requirements analysis, system architecture design, and technical system design to form unified system capability requirements. The task of top-level design is to decompose the capability requirements into system functional requirements, which can be the input of system development at the system level.

At the integration verification layer, integration, joint testing, and capability evaluation are organized after the delivery of the system layer.

At the system level, with product realization as the goal, it mainly includes system requirements analysis, system design, subsystem design, key technology research, subsystem integration test, and system delivery. The system engineering method is used to guide engineering practice at this level.

In terms of organizational structure, according to the idea of hierarchical organization and joint management, engineering activities are organized and implemented by a joint system composed of the overall unit and the undertaking units of each subsystem. Among them, engineering activities at the SoS level are implemented by the overall units, and engineering activities at the system level are implemented by the system undertaking units.

3 Risk transmission mechanism analysis

3.1 Risk analysis

Risk refers to a measure of the impact of uncertainty on project development goals under the constraints of prescribed technology, schedule, and cost. Among them, uncertainty refers to the incomplete understanding of risk sources and their possibilities or consequences; development goals refer to technical performance, schedule, cost, health and safety, environment, products, processes, etc.; impact refers to deviations from expectations, where deviations may include positive or negative results. The risk referred to in this article only focuses on situations that are likely to produce negative results.

Risk usually consists of risk source, possibility (or probability), and consequence (impact). Among them, the existence of a source of risk leads to the existence of the risk, and the risk source is the root cause of the risk. Risks may arise at any stage of engineering development, and they objectively exist for any project, such as technical risks, cost risks, schedule risks, etc., which cannot be completely eliminated. Risks can only be reduced to a minimum through risk control.

3.2 Risk transmission analysis

Risk management is one of the core aspects of project management. It is necessary to identify potential risks during the entire project life cycle. According to the organizational form, logical function, and physical structure, the structure of major scientific and technological projects is divided into different levels such as SoS level, system level, sub-system level, equipment level, module level, and components level. As shown in Fig. 2, risks are analyzed and evaluated from different levels.

Fig. 2

Architecture hierarchy.

As shown in Fig. 2, as a SoS, a major science and technology project is composed of multiple, complex, and giant systems. Each system is composed of multiple sub-systems with different functions. The sub-systems are composed of equipment with different functions. The equipment is composed of different modules. At the bottom layer, the modules are composed of components. In the architecture diagram, SoS is divided into different levels according to the degree of integration; each level is composed of system elements to form a network structure. The elements are related to and influence each other. Simultaneously, risk factors are transmitted and coupled to each other among the elements of the same level. Elements that have a subordination relationship are also related to and affect each other, so that risk factors are transferred and coupled between different levels. Therefore, in the constructed architecture view, the propagation mechanism of risk factors in the entire life cycle of the project can be clearly perceived.

The risk transmission of major scientific and technological projects means that potential risk nodes in different levels of the system become actual risk events in the process of system engineering development. Risk factors will eventually occur through the interactive coupling and cumulative amplification of nodes at all levels of the system, and catastrophic events may have fatal consequences for the entire SoS.

In the SoS, a series of risk chain reactions occurs in the whole system due to the induction of one or several risk events, which is essentially due to the objective existence of risk sources at all levels of the SoS and the self-organizing characteristics of the SoS. The SoS can enter a self-organized critical state through the cascading reaction of node risks within and between levels. At this moment, the occurrence of a risk event at a certain node may cause the spread, accumulation, and amplification of risks at multiple levels in the SoS, resulting in the entire collapse of the SoS. The forest-fire model introduced in the next section is more conducive to understanding the mechanism of risk transmission.

4 Risk transmission model

4.1 Forest-fire model

In the following, we use the forest-fire model to illustrate the risk transmission mechanism of major scientific and technological project systems. The forest-fire model was originally designed by Malamud et al. [4] to study the causes of forest fires in Yellowstone National Park in the United States. The model was simulated on a computer, assuming that a certain area of the forest was randomly selected and a match was dropped to observe changes in the scope and scale of the fire.

Building a forest-fire model includes three rules: firstly, the forest is composed of trees, and the number of trees increases at a fixed rate; secondly, every once in a while, a tree somewhere in the forest randomly catches fire; thirdly, fire spreads to nearby trees, and once a tree catches fire, it ignites the surrounding trees. If the conditions are right, the fire spreads.

The researchers conducted a series of simulations. In each simulation, the number of burning trees or burning area was counted. Like a real forest, the model showed that there were more small fires than large fires. The forest network in the grid can enter a critical state through its own natural adjustment. A single match may be able to cause a fire of any scale and even destroy the entire forest. Malamud et al. called this the Yellowstone Effect. By studying the frequency of large fires burning in the range of 1 and 10 km², the model provided a power law distribution of forest fires. The study showed that the scale and frequency of 4284 fires within the jurisdiction of the U.S. Fish and Wildlife Service from 1986 to 1995 conformed to a power law distribution: when the area of fire burning doubled, the frequency dropped to 1/2.48. This rule applied to fires with a scale difference of 1 million times. In other words, the frequency of fires on different scales conforms to the power law distribution, which is also called Gutenberg–Richter power law of the ecosystem.

4.2 Risk transmission model

Risk events in engineering projects are just like fire events in the forest. They are unpredictable and spread. Risk events have different levels, may break out at any time, and are contagious. These characteristics are similar to those of forest fires. Therefore, we referred to the forest fire model to construct the risk propagation model for an engineering project.

As shown in Fig. 3, the forest-fire model was used to simulate the risk transmission mechanism of major scientific and technological project construction projects. The left figure shows the initial stage of project development, and the right shows the self-organized criticality of the project. The blue indicates the risk source and the red nodes are affected by risk. There are multiple hidden risk sources in each node at each level of major scientific and technological projects. According to the order of appearance of the risk sources, each risk source is distributed in different development stages of the project.

During the initial state, the number of risk sources contained in each project node is small, the risk level is low, and the possibility of its infection to other project nodes is low. When the engineering project reaches the self-organized critical state, the number of risk sources contained in each project node gradually increases, and the risk level is higher. At this time, the possibility of these risk sources infecting other nodes considerably increases and they are more likely to cause disaster or paralysis of the entire project.

In Fig. 3, if a certain risk event of a node is induced, the risk of that node will be transmitted to other surrounding nodes, which will induce other nodes to have related risk events, and the number of affected nodes is limited. With the implementation of major scientific and technological projects, when the number of risk sources at each node increases to a certain extent and the system reaches a self-organized critical state, even a small risk event at a node can cause a series of catastrophes in the SoS.

Fig. 3

Risk transmission model.

The distribution of risks in major scientific and technological projects exhibits power law characteristics. Assuming that there are N nodes in the system, the total number of outbreaking risks is R₀, and the number of outbreak risks transmitted to i (i = 1, 2, ⋯ N) nodes is R_i, then the relationship between risk scale and frequency is: $\frac{R_{i}}{R_{0}} = {(\frac{i}{N})}^{γ},$

where γ = - log ₂2.48 _.

Assuming that x is the risk transfer scale (the number of risk transferred nodes), referring to the proof in [13]: $x = \frac{{dR}_{i}}{di} = γ \frac{R_{0}}{N} {(\frac{i}{N})}^{γ - 1} = x_{0} {(\frac{i}{N})}^{γ - 1} .$

Because ${(\frac{i}{N})}^{γ - 1}$ is a decreasing function, $x_{0} = γ \frac{R_{0}}{N}$ is the minimum value of x, and: $\frac{i}{N} = {(\frac{x}{x_{0}})}^{1 / (γ - 1)} .$

Assuming that the risk severity is calibrated by the risk transfer scale, the distribution function of risk severity can be obtained. $P (X ⩽ x) = 1 - P (X ⩾ x) = 1 - {(\frac{x}{x_{0}})}^{1 / (γ - 1)}$

Then, the probability density of risk severity is: $p_{x} = - \frac{1}{x_{0} (γ - 1)} {(\frac{x}{x_{0}})}^{1 / (γ - 1)^{- 1}} .$

It can be derived that risk severity and risk probability present a power law relationship. The occurrence frequency of mild risk is greater than that of serious risk, and catastrophic risk events rarely occur.

4.3 Results analysis

Risk is a part of system engineering construction, just as fire is a part of a forest ecosystem. Since 1890, the U.S. Forest Service has adopted a zero tolerance attitude toward fire, trying to put out all fires, which is equivalent to reducing the frequency of fire. An unexpected result of this measure is the ageing of the forest. A large number of dead trees, dead branches, flammable shrubs, and grass have accumulated in protected forest, which has created a hidden danger for subsequent large-scale fire. As a result, forest managers no longer try to control small- and medium-sized fires.

The conclusion of the forest-fire model is obvious: it is necessary to have a certain tolerance for smaller fires (risk events), so that the risks can be reflected and released to avoid a devastating disaster to the entire system. The focus of this study was to analyze the risk of the project, and a technical risk transfer model is introduced referring to the forest-fire model.

We suggest using this idea; to avoid catastrophic risk events, small risks should be tolerated in the engineering process of major scientific and technological projects. If all risk events are suppressed, the hidden risk sources of the corresponding nodes will not become overwhelming, and the risk distribution of each node in the system will change. Once the system enters a self-organized critical state, as the risk sources of each node in the system accumulate, catastrophic events may occur at any time. Therefore, to avoid the risk of causing catastrophic consequences for the project, small risks should be timely released and exposed during the process of project implementation. Advanced management technology is needed to identify and avoid whole-process risk. In addition, project stakeholders are required to share the project risks. Through risk allocation and transfer, benefit and risk sharing can be realized.

5 Conclusion

This paper discussed a system engineering process model of major scientific and technological projects from the perspective of system structure and engineering development. The hierarchical structure of the technical risk of major scientific and technological projects was analyzed, covering the SoS, system, subsystem, equipment, module, and components levels. This paper introduced the forest-fire model to the risk management of scientific research projects for the first time. Our analysis of the results showed that the findings are conducive to the scientific decision-making of major scientific and technological projects, to reducing the occurrence of engineering technical risks from the source, and ensuring implementation of major scientific and technological projects.

References

Shishodia

, Dixit

and Verma

, Project risk analysis based on project characteristics[J], Benchmarking25(3) (2018), 893–918.

Wibowo

and Kochendörfer

, Financial risk analysis of project finance in Indonesian toll roads[J], Journal of Construction Engineering and Management131(9) (2005), 963–972.

Thomas

A.V.

, Kalidindi

S.N.

and Ganesh

L.S.

, Modelling and assessment of critical risks in BOT road projects[J], Construction Management and Economics24(4) (2006), 407–424.

Malamud

B.D.

, Morein

and Turcotte

D.L.

, Forest fires: an example of self-organized critical behavior[J], Science281(5384) (1998), 1840–1842.

Liu

and Sun

F.H.

, Research on the risk assessment method of PPP project based on the improved matter element model[J], 27(2) (2020), 614–624.

Taimur

and Ilan

, Financial market contagion in the Asian crises[C], IMF Working Paper, 1998, 155.

Grimsey

and Lewis

M.K.

, Evaluating the risks of public private partnerships for infrastructure projects[J], International Journal of Project Management20(2) (2002), 107–118.

Lee

, Park

and Shin

J.G.

, Large engineering project risk management using a Bayesian belief network[J], Expert Systems with Applications36(3) (2009), 5880–5887.

Li.

, Research on risk transmission mechanism of supply chain [J],’, s Circulation Economy25(1) (2011), 41–44.

10.

Kaminsky

G.L.

and Reinhart

C.M.

, On crises, contagion, and confusion[J], Journal of International Economics51(1) (2000), 145–168.

11.

Cheng

G.P.

and Qiu

Y.G.

, Research on risk transmission model of supply chain [J], Journal of Wuhan University of Technology (Social science edition)22(2) (2009), 36–41.

12.

Yang

and Wang

J.J.

, Risk transfer process of urban utility tunnel PPP project based on complex network model [J], Journal of Zhejiang University (Engineering edition)54(9) (2020), 1666–1676.

13.

Hardy

, Pareto’s law[J], The Mathematical Intelligencer32(3) (2010), 38–43.

14.

Yaniv

, Project risk optimization based on strategy proof risk assessment[D], Massachusetts Institute of Technology, Cambridge, 2010.

15.

Xiang

P.CH.

and Luo

Y.P.

, Study on risk transmission mechanism of social stability in major project construction [J], World Science and Technology Research and Development36(4) (2014), 420–425.

16.

Cui

, Yang

and Shao

X.J.

, Reviewing the transmission mechanism of enterprise risk from the Asian financial crisis [J], Southern Finance2001(10), 27–29.

17.

Shao

and Dong

, Study on the risk of scientific research project based on AHP (Analytic hierarchy process)[J], Shanxi Province and Technology, 2014.

18.

JE Group, Guidance for transportation project management[M], Washington, DC: The National Academies Press, 2009. https://doi.org/10.17226/23028.

19.

Shane

J.S.

, Strong

K.C.

and Gransberg

D.D.

, Project management strategies for complex projects[M], Traansportation Reseach Board. Washington, D.C. 2014

20.

Leng

, Zhang

, Gong

, Zhou

and Liu

, Research on risk management of scientific research projects[C], 10th International Conference on E-Business, Management and Economics, (ICEME 2019), 226–228.

21.

Zhu

J.Y.

and Zhu

S.Z.

, Analysis of transmission mechanism and defense countermeasures of financial risk [J], Journal of Donghua University2001(10), 16–20.

22.

Tang

and Cheng

, Research on risk transfer model of real estate projects[M], Recent Trends in Decision Science and Management, 2020.

23.

Chen

Y.S.

, Chuang

H.M.

, Sangaiah

A.K.

, Lin

C.K.

and Huang

W.B.

, A study for project risk management using an advanced mcdm-based dematel-anp approach[J], Journal of Ambient Intelligence and Humanized Computing10(7) (2019), 2669–2681.

24.

Ren

, Wang

C.C.

, Shen

M.S.

, Sun

X.F.

and Guo

, Research on risk management of major scientific research projects based on improved hall three-dimensional structure[J], Technology and Innovation Management, 2019.