Analysis of the cost factors on E-government software cost using fuzzy decision making system

Abstract

The reasonable cost budget of the e-government scheme can effectively promote the construction of the digital government. To analyze the cost impact components of the e-government system and find out the impact factor model works in China, this paper reviews relevant literature on software cost impact factors and proposes the impact factors model based on COCOMO II. Besides, combined with the actual construction of digital government and specific cases, this paper analyzes the mechanism of each impact factor in detail. The model can be used to guide the cost estimation of e-government software in China, especially with artificial intelligence estimation method. An enhanced decision theory of theory based on fuzzy set has been adopted for analysis of cost factor on E-government software cost.

Keywords

E-government information system software cost estimation impact factors fuzzy set

1 Introduction

E-government is usually defined as the management and service functions are integrated by using information technology based on the information network as the platform. Government departments at all levels can optimize and reorganize the organizational structure and working process of the government, and provide high-quality, standardized, and transparent public services, to realize the integrated management and operation of the affairs of the public, government, business and service [1]. The realization form of E-government is the E-government system, which belongs to the category of information system. In a broad sense, the E-government system includes network platform, storage platform, application platform, hardware equipment, program system, human and many other components, which is a system with complex network and three-dimensional cross structure. In the narrow sense, the E-government system refers to the program system itself [2]. At present, the construction of the E-government system is very common, and the scale of investment is getting larger and larger. However, the measurement of information system cost lacks a scientific and effective standard. Therefore, the budget of the E-government information system cost has always been one of the difficult problems for government organs to carry out information construction. It is very important to measure the cost scientifically, which is related to the development of business.

The Information system is usually a man-machine system with the large scale and high complexity, the process of development, use, maintenance, and management is a complex system engineering. The costs can be divided into development costs and operational maintenance costs. The former includes the hardware cost, the system software cost, and the development cost of application software [3]. However, the development cost of application software is closely related to personnel quality, experience, familiarity with system software, and user intervention at any time, which is difficult to predict accurately. The software cost calculation is a scheme or model utilized to understand the development workload, duration, and cost of a software models, which applies to the budget, bidding, project planning, change, settlement, audit, or transaction of a software project [4], Overview diagram of cost estimation is shown in Fig. 1. A reasonable software cost evaluation method can improve the accuracy of the software budget, guarantee the delivery cycle of software projects, reasonably arrange and schedule the research personnel, and thus improve the quality of software projects. However, estimation of software cost is a very complex task, and the final cost of software development is influenced by many factors, many of which are difficult to grasp. Therefore, it is difficult to have an estimation method and estimation tool suitable for various software development [5]. The effective estimation for different project development environments has become a difficult problem in the field of software cost estimation [6]. Therefore, it is of great scientific significance and practical value to estimate the components affecting information scenario cost and select the appropriate method for estimation before adopting the technology of the software cost evaluation.

Fig. 1

Overview diagram.

Fuzzy set can be understood as an inclusion and overall simplification of classical groups. It can be dealt with the setting of set membership. Normally it allows limited membership which deals with elements that have different membership degrees in the group. Initially from this we could handle the variations amongst classical group and fuzzy group. Classical set includes parameters that satisfy exact properties of membership while fuzzy group includes elements that content inexact membership properties.

In the field of investigation on the impact factors of E-government information cost, no one has explicitly proposed the impact factors model for China [7]. In order to promote the development of business, it is required to study the effecting factors model of the cost of E-government system. By searching for the relevant information, this paper summarizes the impact factors of the information system cost and the characteristics of E-government. Combining with the specific cases and the construction of E-government system of Shandong province in China, this paper analyzes the impact factors of the cost of China’s E-administration information scenario, and determine the 10 indicators, including the level of software classified protection, software reliability, software complexity, software reuse, project location, difficulty of software platform, software development environment, schedule target, policy compliance, characteristic project.

The contents of other parts of this paper are as follows: Section 2 summarizes the cost estimation methods and impact factors of the information system. According to the impact factors of the COCOMOII system and the characteristics and development requirements of E-government, the impact factors of E-government system cost are proposed in Section 3. Based on the construction of the E-administration system in Shandong province, the impact factors are analyzed in Section 4. Section 5 carries on the conclusion and the prospect of this paper.

2 Related work

As a kind of material system, the E-government scenario has the general rule of the information system. The cost estimation of the information system mainly lies in the calculation of the development cost of the application. After decades of forming, the software cost estimation method has been gradually improved in technology and effect, which is mainly divided into the following methods: Algorithmic Methods, Learning-Oriented Methods, Analogy, and Expert Judgment. The Algorithmic Methods include the COCOMO model, Function point Analysis, Use case point Analysis, etc [8 –10]. The Learning-Oriented Methods includes neural network, genetic algorithm [11, 12], heuristic algorithm, decision tree, fuzzy logic, etc. The Analogy can be divided into the analogy method, analogy method, and cluster analysis method [13].

Among them, The Expert Judgment can reconcile the opinions of various experts and can be evaluated when the historical data set is missing or insufficient. However, the Expert Judgment is very dependent on the expert’s experience, and the accuracy of software project evaluation depends on the expert’s ability, so the expert decision evaluation method is easily influenced by the expert’s subjective factors. The analogy estimates by treating the attribute similarity between the evaluated project and the existing project. However, the absence of data in the historical data set, the selection of attributes, the definition, and adaptation of similarity, as well as different verification and evaluation criteria will affect the evaluation results of the analog-based method. The Algorithmic Method is logical and simple. However, it is very difficult to determine the calculation formula. Foreign researchers rely on a huge total of software project data to verify and adjust the model. Besides, the parameter value of model evaluation still depends on experience, and the parameter setting is biased to the development of foreign software, which does not apply to the cost estimation of domestic software. The advantage of the Learning-Oriented Method is that the learning and training are spontaneous, the acquisition of knowledge and information is comprehensive, and it has good ability of nonlinear fitting and self-learning. The disadvantage is that a large number of sample sets are required for training, and input parameters are designed artificially. If input parameters cannot describe project information well, the evaluation effect will also be affected [14, 15]. Zhang and Huang [16] have mentioned a fuzzy risk estimation model for detail methodology service subcontracting in which the factors of risk are assessed by utilizing fuzzy range and grades risk of every risk factor which are used for estimating the fuzzy linguistic value. Fuzzy logic and fuzzy group can illustrate uncertain, ambiguous or imprecise details in managing e-government enthusiasm by mentioning inaccuracy in decision taken by human [17]. Those estimation methods and their limitation can be seen in Table 1.

Table 1
Software evaluation methodologies and drawbacks

Evaluation method Drawbacks

Evaluation of Experts Depends on knowledge of experts, perception and vision which can be single handed (certainly forming to overoptimistic calculation)

Unable to quantify and justify all factors applied by experts in the process of estimation

Algorithmic Methods Certainly will not take teams skill set involving in projects, certain projects of organization and management of projects.

Not able to adopt for modern methodologies, programming which includes codeless and development of agile.

Learning-Oriented Methods Can find from earlier data and forecast the outcome of future based on data.

A large number of sample sets are required for training, and input parameters are designed artificially.

Similarity Choice of performance of associating criterion and logics of variance findings (phases of confidence).

Analogous requirement of models for associating which is barely able to achieve in development of software.

It can be seen from the above that the input impact factors are very critical in most estimation methods of software cost, but according to the existing literature, does not involve relevant scheme on impact components of E- administration. Due to the time and space differences between the estimated project and the historical project data in different regions and fields, it is necessary to analyze the impact factors of the current project based on a flexible and perfect impact factors system. Similar to it is the impact factor system proposed based on the Algorithmic Methods. Among them, the advantage of the COCOMO model method is that the factors are more detailed and all factors affecting workload are classified and graded, including product component, platform scheme, personnel factors, and project methodologies, so it is easy to calculate [16]. It is often possible to remove factors that have local meaning only in certain situations, thereby greatly reducing the impact of project differences and making the selected factors applicable to most software projects. According to the actual situation of the enhancement process of China’s E-administrative model, some cost drivers can be appropriately added, deleted or merged, to find the universal impact factor system which is suitable for China’s E-government system.

3 Result

This part mainly discusses the construction process of the impact component of the E-government model cost. That is, founded on the early cost-driven factor structure proposed by COCOMO II, the characteristics of E-government are summarized by sorting out a large number of literature and combining with the actual review of the E-government detailed model. Meanwhile, the impact issues of the E-government system cost was constructed based on the modification principle of COCOMO II.

3.1 Determine the research object

According to the growth stage of the E-government system, the cost is divided into the development cost and the operational maintenance costs cost. The cost classification method consistent with the growth stage of the information system, and it can adopt the development norms of the international popular information system and the relevant national information standards, to facilitate the cost accounting and analysis according to the needs of the system development at the different times and levels. In terms of the components, the cost consists of materials, water and electricity of infrastructure, labor and training, management, and other expenses. The different cost objects have different predictability, such as the cost of hardware facilities, system software, etc. can be calculated with the current market price as a reference after the system planning and design is completed. On the other hand, the development cost of application software is closely related to personnel quality, experience, familiarity with system software, and user’s intervention at any time and other factors that are difficult to measure, especially in the early planning stage [17], which is difficult to predict accurately. At the same time, the most important thing is to budget the cost before the project bidding in the actual construction of the E-government system. The results will affect the progress and excellence of the following project construction. Therefore, the thing of this paper is the impact factors of software price in the early stage of E-government.

3.2 The impact of the COCOMO II scheme

The rate driver is portioned into four categories in COCOMO II. The third category is Platform attributes, including Time Constraint, Storage Constraint, Platform Volatility. The fourth category is the Project attributes, including Use of Software Tools, Multisite Development, Required Development Schedule. The post architecture of COCOMO II and the early design of COCOMO II can be seen in Tables 2 and 3. Due to the uncertainty in the early stage, the cost drivers in the above four categories were simplified and finally adjusted to seven indexes, including Product Reliability and Complexity(RCPX), Reusability(RUSE), Platform Difficulty(PDIF), Personnel Capability(PERS), Personnel Experience(PREX), Facilities(FCIL), and Required Development Schedule (SCED) [18]. This paper would adjust the early cost drivers of COCOMO II.

Table 2
The post architecture of COCOMO II

Cost drives Introductions Cost drives Introductions

Product Factors RELY Required Software Reliability Personnel Factors ACAP Analyst Capability

DATA Database Size AEXP Applications Experience

DOCU Documentation Match to Life-Cycle Needs PCAP Programmer Capability

CPLX Software Product Complexity PEXP Platform Experience

RUSE Required Reusability LTEX Language and Tool Experience

PCON Personnel Continuity

Platform Factors TIME Execution Time Constraint Project Factors TOOL Use of Software Tools

STOR Main Storage Constraint SCED Required Development Schedule

PVOL Platform Volatility SITE Multisite Development

Cost drives	Introductions	Cost drives	Introductions
Product Factors	RELY	Required Software Reliability	Personnel Factors	ACAP	Analyst Capability
	DATA	Database Size		AEXP	Applications Experience
	DOCU	Documentation Match to Life-Cycle Needs		PCAP	Programmer Capability
	CPLX	Software Product Complexity		PEXP	Platform Experience
	RUSE	Required Reusability		LTEX	Language and Tool Experience
				PCON	Personnel Continuity
Platform Factors	TIME	Execution Time Constraint	Project Factors	TOOL	Use of Software Tools
	STOR	Main Storage Constraint		SCED	Required Development Schedule
	PVOL	Platform Volatility		SITE	Multisite Development

Table 3

The early design of COCOMO II

Cost drives	Introductions	Cost drives	Introductions
RCPX	Product Reliability and Complexity	PREX	Personnel Experience
RUSE	Required Reusability	FCIL	Facilities
PDIF	Platform Difficulty	SCED	Required Development Schedule
PERS	Personnel Capability

3.3 The characteristics of E-government

The E-government information system has its characteristics. With “E-government” and “cost” as the search terms, the author obtained 245 articles from the CNKI (China’s most comprehensive database of Chinese periodicals) and ‘Web of Science’ between 2015 to 2020 which is taken as the research objects and summarized the relevant contents described in the literature. Besides, the author’s institution is responsible for the review of E-management material system in Shandong province. Therefore, the author summarizes the following development characteristics of E-government based on theory and practice in the process of work. (1) It has become an international mainstream trend to build E-government with public service as the core [19]. As early as March 2013, the Department of Economic and Social Affairs of the United Nations released the “2012 United Nations E-government survey report: E-government for the public”, with the theme of “E-government for the public’’, in which the development concept of “building an excellent government service culture guided by public needs” was put forward. One of the main goals of E-government development is to improve the level of public services and provide quality and efficient basic public services to the grassroots. (2) We should improve the security of the E-government service system and promote the sustainable development of E-government [20]. The E-government includes economic, political, and military information, and its security determines whether the administrative affairs of government agencies can be carried out normally. Therefore, information security is a prerequisite for the construction and normal operation of E-government, which is of far-reaching significance for promoting the healthy and orderly development of E-government in China. (3) Promote the circulation and sharing of government service data to break down information barriers [21, 22]. In the digital government environment, the circulation and sharing of government service data can effectively promote the business restructuring and process reengineering of government service, build the government service chain, and improve the government governance capacity. (4) Strengthen innovative development, and transform from passive response to active leadership. With the transformation of government functions and the practical requirements of the reform of “Streamline administration, delegate powers, and improve regulation and services”, we will strengthen the organizational innovation, model innovation, technological innovation, and application innovation, and use concept innovation to lead the new model of future development, to enhance enormous development ability of E-government in China.

3.4 Principles of modification

When revising the cost driver, we mainly improve it according to the following principles. (1)Definability principle, that is, the parameters determined can be clearly defined, and abstract evaluation needs to be converted into concrete numbers for calculation.(2) Applicability principle, that is, we will observe whether this index plays a significant role in China’s E-government information system, as well as the influence of factors in China’s software industry.(3) Materiality principle, that is, the choice of cost drivers focuses on factors that have a significant impact on software development, and that have an impact on all projects. (4) The influence principle of the software division, that is, the software under different types and scales should consider different factors in the cost estimation. Therefore, software types, sizes, and correction methods should be combined when revising the driving factors of costs. (5) The practical principle, that is, when analyzing and revising the driving factors of software cost, we mainly analyze based on the characteristics of China’s current E-government and apply on basis of basic correction method associated with the characteristics of E-government [24].

3.5 System and explanation of impact factors of the cost

From a system perspective, the characteristics of E-government are integrated into the early cost driver system of COCOMO II. The 10 impact factors are proposed, including policy compliance, level of software classified protection, project location and featured project, the difficulty of software platform, software reliability, software complexity, function reuse, development environment, and schedule requirements. Among them, the difficulty of software platform, software reliability, function reuse, development environment, and schedule requirements are adjusted based on the COCOMO II, while the policy compliance, level of software classified protection, featured project and project location are newly added impact factors according to the characteristics of the E-government system. The relationship between COCOMO II model and ours is shown in Fig. 2.

Fig. 2

The impact factor model of E-government information system cost.

4 Discussion

This part will combine the specific case to discuss the explanation of each impact factor and its relationship with the E-government system cost. Based on the above content, this part will makes an empirical analysis of each influencing factors which can divided into the retained cost impact factor, added cost impact factor, and deleted cost impact factor, and draws the box diagram for the impact factors and the final cost of through the actual e-government system data, which can be seen in Fig. 4.

Fig. 3

Trapezoidal fuzzy number P.

Fig. 4

The impact of variables on software cost.

4.1 Retained cost impact factor

The software reliability is derived from the reliability and complexity indexes of early cost drivers in COCOMO II, and also incorporates the characteristics of public service in E-government. In COCOMOII, software reliability is an estimation of further performance of software functionalities over certain time, and the value is determined by the impact of failure. For the E-government, the main service object is the government departments and the public. People’s demand for public service quality is growing day by day, and higher availability are put forth for the government’s service capacity [25]. This index is evaluated primarily by whether the object orientation is public or government, and how often it is used.

The software complexity is derived from the reliability and complexity indexes of early cost drivers in COCOMO II, and also incorporates the characteristics of information sharing in E-government. In COCOMO II. For the E-government, the construction of information resources sharing of E-government involves all aspects of the society, various fields, and departments. It is a complex system engineering, which relies on overall planning and cooperation among shared subjects [26]. This index is mainly evaluated by whether vertical and non-vertical sectors exchange data with other sectors, and whether financial or quasi-financial situations are involved. The software complexity has a significant signs on the E-management information scheme cost.

The function reuse is derived from the reusable development indexes of early cost motorists in COCOMO II, which describes the additional effort required to construct components that can be reused in past or the future projects. Exertion is observed in forming more certain software designs, detailed documentation and testing effortlessly to make the component can be used in other applications. This index is mainly evaluated through phase I, phase II, and phase III and above. Functional reuse has a negative impact on the E-manegement information system cost.

The difficulty of software platform is derived from the difficulty of platform indexes of early cost drivers in COCOMO II. The platform attribute refers to the influence of platform factors on software workload measurement from the three following aspects: the execution time limit, main storage limit, and platform change rate. The platform refers to the operating environment of hardware and software during the project. E-management details scheme is mainly built in the Internet area, administrative service area, and E-government extranet [27, 28]. This index is evaluated through different construction conditions. The difficulty of software platform does significant improvement on the E-management information scheme cost.

The development environment is derived from the project facilities indexes of early cost drivers in COCOMO II. The project facilities refer to software development tools and multi-point development [23]. For the early E-government information system, there is no specific development link, so it is mainly evaluated by the development language and information system form.

The schedule requirement is derived from the development schedule required indexes of early cost drivers in COCOMO II, which measure the schedule constraints imposed on software development projects. The rating of these indexes is based on the percentage of extension or acceleration in nominal progress relative to projects requiring a certain amount of work. The schedule acceleration tends to require more work in the early stages to eliminate risk and refine the architecture. The schedule requirement is mainly evaluated by the number of actual application months, which has a positive impact on the E-government information system cost.

4.2 Fuzzy set approach

The theory of fuzzy set is modelled to evaluate utilizing mathematically illustrating vagueness and uncertainty and gives tools which are pompous for managing with inconsistent inherent to issues related to decision forming. A fuzzy number includes a subset of fuzzy in the universe of dissertation A that is included with normal and convex. Normally, different varieties of fuzzy numbers like trapezoidal, triangular, bell- shaped numbers are utilized in making decisions. Normally, numbers in trapezoidal are broadly utilized due to basic mathematical illustration and easy estimation. A trapezoidal fuzzy total form (x, y, u, v) is vital generic component class of fuzzy total with membership linear function. Due to generic suitably of this fuzzy number class, it identifies in displaying and practical engineering issues rather than using triangular fuzzy total.

4.3 System of comprehensive trapezoidal fuzzy numbers

A normalized trapezoidal fuzzy total can be definite as P = (p₁, p₂, p₃, p₄ ; S_P) , and the function of membership μ_P (a) : K → [0, 1] is known as given below: $μ_{P} (a) = {\begin{matrix} \frac{a - p_{1}}{p_{2} - p_{1}} \times S_{P}, a \in (p_{1}, p_{2}) \\ \frac{a - p_{4}}{p_{3} - p_{4}} \times S_{P}, a \in (p_{3}, p_{4}) \\ 0, a \in (- \infty, p_{1}) \cup (p_{4}, \infty) \end{matrix}$ (1)

Here, p₁ ⩽ p₂ ⩽ p₃ ⩽ p₄ and S_P ∈ [0, 1]. The basics of the normally trapezoidal fuzzy total a ∈ K are number real, and its function of membership μ_P (a) is continuous and regular convex of function, it depicts the degree of membership to the fuzzy sets. If -1 ⩽ p₁ ⩽ p₂ ⩽ p₃ ⩽ p₄ ⩽ 1, then P is known the normalized fuzzy number in trapezoidal model. Normally, if S_P = 1, then P is known as the trapezoidal fuzzy total (p₁, p₂, p₃, p₄) ; if p₁ < p₂ = p₃ < p₄, then P is minimized to a triangular fuzzy total. If p₁ = p₂ = p₃ = p₄, then P is minimized to a real number.

4.4 Added cost impact factor

The level of software classified protection is derived from the security requirements of E-government development. National information security involves the security of national defense, economy, culture, and politics. Therefore, only when the level of E-government information security is raised to a certain height can the national security barrier be guaranteed [23]. For the early E-government information system, the solution to the systemic security problem of the current E-government is active defense and repair after being violated. This index is mainly evaluated by the level of software classified protection, which has a positive impact on the E-government information system cost.

The policy compliance is a new index. The ability of provincial governments to “Internet+government service” is affected by multiple and complex factors, among which the basis of E-government and the importance of leaders are the key factors [29, 30]. The development of E-government is inseparable from the government’s attention and planning. As an important carrier, the E-government system needs to combine policy and planning, which determines the quality of its construction. This index is mainly evaluated through the attention of leaders and included in the planning of departments at all levels and their departments, which has optimistic impact on the E-government information system cost.

Whether the featured project is derived from the application innovation in the growth requirements of E-management. As an important part of informatization, the new information technologies represented by big data, cloud computing, block chain, and artificial intelligence will play a more important role in future E-government services. This will promote the integration of government resources and optimize the process of government affairs, for quality improvement and efficiency of government utilities and better promote the modernization of the country’s governance system and governance capacity. This index is mainly evaluated by the integration of new information technology to achieve the innovation of governance and service, which has optimistic deviation on the E-government information system cost.

The project location is the new index. The gap between China’s less developed areas and the southeast coastal areas includes the dissemination, acquisition, and application of knowledge and information. In the government system, it is manifested as the “Digital gap” in the construction and application of E-government. The technology and method of government management in the less developed areas are lagging behind those in rich areas. And less-developed areas stagnate in the government information, resulting in a further widening of the gap between the levels of government efficiency. Therefore, the construction of the E-government information system needs to be carried out according to the conditions in different regions.

4.5 Deleted cost impact factor

Here we have discussed the estimation of the E-government information system cost in the early planning stage. Before the bidding stage of the project. Since it was before the bidding stage of the project, the impact factors of developers and implementation processes were not involved in the project review process, so the two impact factors of the early cost drivers in COCOMO II, namely, personnel capability and personnel experience, were excluded.

4.6 Summary

In the 10 cost impact factors, the evaluation content of the retained cost impact factor considers the characteristics of E-government, such as software complexity and software reliability. At the same time, combined with the actual construction of E-government, the project location and policy compliance are added. Combined with the actual project review process, the personnel capability and personnel experience were deleted. In the specific application process, it is recommended to apply COCOMO II and artificial intelligence estimation method that is Learning-Oriented Methods. The former is assigned by its defined criteria, and the corresponding parameters are adjusted to improve the accuracy. The latter requires training through a large dataset of related projects.

5 Conclusion

Although there are many problems and difficulties, it is very significant to study the impact factors of the E-management system cost for its development. With the development of AI, it is playing more important role in E-government system cost estimation, by determining the cost impact factors, the corresponding cost evaluation technologyand, in particular, artificial intelligence estimation method, can be deployed more accurately, to improve the accuracy of project investment and the efficiency of construction. This paper sorts out the relationship and analyzes the development status of E-government in China, and the cost evaluation of information systems and establishes the model of impact factors of E-government information system cost in China. The study explodes to utilize the trapezoidal fuzzy number for observing linguistic illustration of cost assessment. At the same time, combining with the actual construction situation and specific cases of the digital government, this paper analyzes the impact factors, which lays a foundation for the cost estimation of the E-government system in the future.

In future work, we will use more iterative of sample data and open models to modify and improve this model, determine the factors that affect China’s E-government information system, and promote the healthy development of China’s E-government.

Footnotes

Acknowledgment

This work was supported by Key Research and Development Program of Shandong, China (No. 2019GGX105020), and Plan of Youth Innovation Team Development of colleges and universities in Shandong Province (SD2019-161).

References

Ahmad

, Omar

, Azman Yasin

H.I.

, et al., A measurement model of information sharing in E-government services: A case study of Kedah, Malaysia[J]. 2018.

Alharbi

, Papadaki

and Dowland

, The impact of security and its antecedents in behaviour intention of using e-government services[J], Behaviour & Information Technology 36(6) (2017), 620–636.

Barry

, Software engineering economics[J], New York, 1981, 197.

Boehm

, Clark

, Horowitz

, et al., Cost models for future software life cycle processes: COCOMO 2.0[J], Annals of Software Engineering 1(1) (1995), 57–94.

Chen

Y.C.

, Hu

L.T.

, Tseng

K.C.

, et al., Cross-boundary e-government systems: Determinants of performance[J], Government Information Quarterly 36(3) (2019), 449–459.

Dos Santos

L.P.

and Ferreira

M.G.V.

, Safety critical software effort estimation using COCOMO II: A case study in aeronautical industry[J], IEEE Latin America Transactions 16(7) (2018), 2069–2078.

Ejdys

, Ginevicius

, Rozsa

, et al., The role of perceived risk and security level in building trust in E-government solutions[J]. 2019.

Gharehchopogh

F.S.

and Pourali

, A new approach based on continuous genetic algorithm in software cost estimation[J], J. Sci. Res. Dev 2(4) (2015), 87–94.

Karner

, Resource estimation for objectory projects[J], Objective Systems SF AB 17 (1993), 1–9.

10.

Langsari

and Sarno

, Optimizing effort and time parameters of COCOMO II estimation using fuzzymulti-objective PSO[C]//2017 4th International Conference on Electrical Engineering, Computer Science and Informatics (EECSI), IEEE (2017), 1–6.

11.

Lee

S.Y.

and Yoon

H.J.

, A study on the 4th industrial revolution and e-government security strategy-in terms of the cyber security technology of intelligent government[J], The Journal of the Korea Institute of Electronic Communication Sciences 14(2) (2019), 369–376.

12.

and Shang

, Service quality, perceived value, and citizens’ continuous-use intention regarding e-government: Empirical evidence from China[J], Information & Management 57(3) (2020), 103197.

13.

Lin

, Government Management Model of Non-profit Organizations Based on E-government[C]//Proceedings of the 2019 7th International Conference on Computer and Communications Management (2019), 164–168.

14.

Collan

, Fulle'r

and Mezei

, A fuzzy pay-off method for real option valuation, Journal of Applied Mathematics and Decision Sciences, 2009.

15.

, Digital divide and citizen use of e-government in China’s municipalities[J], International Journal of Public Administration in the Digital Age (IJPADA) 5(3) (2018), 16–31.

16.

Mahmoodi

R.K.

and Nojedeh

S.H.

, Investigating the effectiveness of e-government establishment in government organizations[J], Procedia-Social and Behavioral Sciences 230 (2016), 136–141.

17.

Mittas

and Angelis

, LSEbA: least squares regression and estimation by analogy in a semi-parametric model for software cost estimation[J], Empirical Software Engineering 15(5) (2010), 523–555.

18.

Pospieszny

, Czarnacka-Chrobot

and Kobylinski

, An effective approach for software project effort and duration estimation with machine learning algorithms[J], Journal of Systems and Software 137 (2018), 184–196.

19.

Prokopova

, Silhavy

and Silhavy

, Analysis of the Software Project Estimation Process: A Case Study[C]//Computer Science On-line Conference. Springer, Cham, 2019:456–467.

20.

Putra

D.A.D.

, Jasmi

K.A.

, Basiron

, et al., Tactical steps for e-government development[J], International Journal of Pure and Applied Mathematics 119(15) (2018), 2251–2258.

21.

Rad

M.S.

, Nilashi

and Dahlan

H.M.

, Information technology adoption: a review of the literature and classification[J], Universal Access in the Information Society 17(2) (2018), 361–390.

22.

Rashid

, Nisar

M.W.

, Mahmood

, et al., Study of software development cost estimation techniques and models[J], Mehran University Research Journal of Engineering & Technology 39(2) (2020), 413.

23.

Ngan

Roan Thi

, Ali

Mumtaz

, Fujita

Hamido

, Giang

Nguyen Long

, Manogaran

Gunasekaran

and Priyan

, “A New Representation of Intuitionistic Fuzzy Systems and Their Applications in Critical Decision Making’’, IEEE Intelligent Systems

24.

Saeed

, Butt

W.H.

, Kazmi

, et al., Survey of software development effort estimation techniques[C]//Proceedings of the 2018 7th International Conference on Software and Computer Applications (2018), 82–86.

25.

Sarno

and Sidabutar

, Comparison of different neural network architectures for software cost estimation[C]//2015 International Conference on Computer, Control, Informatics and its Applications (IC3INA). IEEE (2015), 68–73.

26.

Wang

Shangguang

, Zhou

, Yang

Mingzhe

, Sun

Lei

and Hsu

Ching-Hsien

, “Service Composition in Cyber-Physical-Social Systems’’, IEEE Transactions on Emerging Topics in Computing.

27.

Stefanovic

, Marjanovic

, Delić

, et al., Assessing the success of e-government systems: An employee perspective[J], Information & Management 53(6) (2016), 717–726.

28.

Symons

C.R.

, Function point analysis: difficulties and improvements[J], IEEE Transactions on Software Engineering 14(1) (1988), 2–11.

29.

Usharani

, Ananth

V.V.

and Velmurugan

, A survey on software effort estimation[C]//2016 International Conference on Electrical, Electronics, and Optimization Techniques (ICEEOT), IEEE (2016), 505–509.

30.

Velarde

, Santiesteban

, Garcia

, et al., Analyzing the effect of variables in the software development effort estimation[J], IEEE Latin America Transactions 14(8) (2016), 3797–3803.

31.

, Ming

and Chen

, Research on government service data sharing in the environment of digital government[J], Administrative Tribune (1) (2018), 8.

32.

Zhang

and Huang

, Fuzzy risk evaluation method for information technology service outsourcing, Journal of Convergence Information Technology 7 (2012), 611–617.

Evaluation method	Drawbacks
Evaluation of Experts	Depends on knowledge of experts, perception and vision which can be single handed (certainly forming to overoptimistic calculation)
	Unable to quantify and justify all factors applied by experts in the process of estimation
Algorithmic Methods	Certainly will not take teams skill set involving in projects, certain projects of organization and management of projects.
	Not able to adopt for modern methodologies, programming which includes codeless and development of agile.
Learning-Oriented Methods	Can find from earlier data and forecast the outcome of future based on data.
	A large number of sample sets are required for training, and input parameters are designed artificially.
Similarity	Choice of performance of associating criterion and logics of variance findings (phases of confidence).
	Analogous requirement of models for associating which is barely able to achieve in development of software.