New Quality Function Deployment Integrated Methodology for Design of Big Data E-Government System in Egypt Ahmed Ahmed Attia PHd, Space Projects Quality Expert. Lecturer of MIS, Modern Academy, Cairo, Egypt aattia51@gmail.com Hesham Mahmoud Ibrahim PHd, Lecturer of MIS, Modern Academy, Cairo, Egypt dr.heshammahmoud@yahoo.com Abstract QFD was developed to transform the voice of customer into designs. Once you have identified a set of software requirements, you will usually want to prioritize them. This study addresses a QFD integrated methodology for improvement of the Egyptian E-Government system by using the Analytical Hierarchy Process (AHP) to prioritize citizen requirements, Quality Function Deployment, to trace weighted customer requirements and quality requirements through all phases of design, and Theory of Inventive Problem Solving (TRIZ) to solve for the contradictions among requirements using the contradiction matrix and the inventive principles. Keywords: Quality Function Deployment; Analytical Hierarchy Process; TRIZ, House of Quality; E-Government; Contradiction Matrix; Voice of customer. 1. Introduction The Analytical Hierarchy Process (AHP) was developed by Thomas Saaty and applied to software engineering by Joachim Karlsson and Kevin Ryan in 1997. AHP is a method for decision making in situations where multiple objectives are present. This method uses a pair-wise comparison matrix to calculate the relative preference, value or cost of software requirements. By using AHP, the requirements engineer can also confirm the consistency of the result. It should be borne in mind that QFD, and any other quality tool or technique, should not normally be used in isolation, but should be an integral part of the continuous improvement process. The union of QFD with other methods will yield even greater benefits from its applications. Integrated QFD is an appropriate framework for integrating other principles and methods (like TRIZ and Taguchi and other methods), where the voice of the customer (client) guides the application of efforts to eliminate waste and foster continuous improvement. Figure 1 illustrates the linkage among QFD, TRIZ and Taguchi methods in this paper, I used only the connection between QFD and TRIZ. Integrating QFD with Taguchi methods is left to future work. Quality Function Deployment (QFD): is a methodology that incorporates the voice of the customer (Customer Requirements) into the design of the product or service QFD provides the customer input TRIZ: is to find a design solution for the contradicting requirements by using the Contradiction Matrix and the 40 inventive principles Taguchi: provides the process for determining the best parameter values for a robust design - Taguchi s methods lack the customer-driven priorities (provided by QFD) and system definition (provided by TRIZ) The linkage can recycle through the sequence of QFD, TRIZ, Taguchi several times (as shown in figure 1). The Different classifications of QFD are shown in Figure 2, and in figure 3, the general view of the House of Quality (HoQ) in QFD is shown. The House of Quality (HoQ) has been automated using an Excel worksheet in order to save time and effort and to allow for easy modifications. Theory of inventive problem solving (TRIZ) was originated by the Russian scientist and engineer Genrich Altshuller. One of the important uses of TRIZ is to find a design solution for the contradicting requirements by using the Contradiction Matrix and the 40 inventive principles. Figure 1: The QFD, TRIZ and Taguchi Connection: Customer-Driven Robust Innovation ASE 2014 ISBN: 978-1-62561-003-4 1
electronic transactions with citizens and businesses can be successful. There are a number of forces motivating G2G e-government initiatives. One of these involves legislation. A second is the interest in improved efficiency. A third is the growing attention being paid to improving the management of government information technology and public resources [2]. Figure 2: A classification for the applications of QFD 2.2 Government-to-Business (G2B) Government-to-Business (G2B) initiatives receive a significant amount of attention, in part because of the high enthusiasm of the business sector and the potential for reducing costs through improved procurement practices and increased competition. The G2B sector includes both the sale of surplus government goods to the public, as well as the procurement of goods and services. There are two primary forces driving the G2B sector. The first is the business community. The second is the growing demand by policymakers for cost cutting and more efficient procurement. 2.3 Government-to-Citizen (G2C) The third e-government sector is Government-to-Citizen (G2C). G2C initiatives are designed to facilitate citizen interaction with government, which is the primary goal of e-government. These initiatives attempt to make transactions, such as renewing licenses and certifications, paying taxes, and applying for benefits, less time consuming and easier to carry out. Interest in G2C initiatives is driven by a combination of several factors. One is citizen demand, G2C initiatives are also driven by an interest in better government through improved efficiency and more reliable outcomes. Figure 3: House of Quality (HoQ) in QFD 2. Main Concepts and Strategy of E-Government System The e-government is summarized as the continuous optimization of service delivery, constituency participation, and governance by transforming internal and external relationships through technology, the Internet, and new media. E-government provides new opportunities to enhance governance, which can include improved efficiency, new services, increased citizen participation, and an enhanced National Information Infrastructure. The goal of digital government is to transform the relationship between government and society in a positive manner. By using ICT, government can modernize public service delivery and promote more citizen engagement in politics. As a consequence, digital government has the potential to change the relationship in such a way that people view government as more accessible, participatory, responsible, transparent, responsive, efficient, and effective than before (see figure 4) [1]. Three distinct sectors can be identified. These include government-to-government (G2G), government-tobusiness (G2B), and government-to-citizen (G2C). 2.1 Government-to-Government (G2G) The G2G sector represents the backbone of e- government. Governments must enhance and update their own internal systems and procedures before Figure 4: E-Government View ASE 2014 ISBN: 978-1-62561-003-4 2
Digital government evolution models were developed a decade ago (see figure 5). They provide guidelines to government organizations on how to incrementally implement digital government. Figure 5: E-Government Model Development Phases The main public services for Citizens are: Income taxes, Job search services, Social security contributions (unemployment benefits family allowances medical costs, etc.), Personal documents (Passports driving licenses), Car registration, Application for building permission, Declaration to the police, Public libraries, Certificates request and delivery, Enrollment in higher education, Announcement for moving (change of address) and Health related services. 2.3.1 Better service: more convenient and reliable Every civil service will be delivered by just a click of a finger. Through the single window on the Internet, online civil applications and document issuance will be available. 2.3.2 The best environment for business By bringing most services for businesses to a single window that integrates services for government procurement, businesses can benefit from the enhanced convenience and transparency. All the procedures from export or import, loading or unloading, customs, to delivery will be brought online. 2.3.3 Productive, Transparent, and Democratic Government The government can work more efficiently thanks to the informatization of core work procedures. The time and money saved by this can further contribute to the improvement of civil services for the people. Real-time open processing of civil applications and a bilateral communication system between the people and the government will develop a more transparent and democratic government. 2.4 Potential Challenges to E-Government An effective strategy will result in significant improvements in the E-government, including: Simplifying delivery of services to citizens; Eliminating layers of government management; Making it possible for citizens, businesses, other levels of government and employees to easily find information and get service from the government; Simplifying agencies' business processes and reducing costs; Streamlining government operations to guarantee rapid response to citizen needs. The E-Government Task Force conducted 71 interviews with more than 150 senior government officials during the process to gather and identify strategic E- Government opportunities [2]. In addition, nearly 200 projects were identified from e-mails sent primarily by federal employees. The overall findings were that agency executives and line professionals want the government to: Use the Web to provide services such as benefits, recreational opportunities, and educational materials; Share information and integrate federal, state and local data where appropriate and possible; Reduce burden on businesses by adopting streamlined processes that promote and enable consolidation in data collection; Adopt commercial best practices to reduce operating costs and make it simpler for government employees to perform their jobs; and Define measures of success and regularly monitor and measure performance. Reducing Overlap and Redundancy to Make It Easier for Citizens to Get Service and to Reduce Costs 2.5 Key Lessons in Developing e-government Systems The following seven recommendations can assist government leaders with building the capabilities and resources to design, implement and maintain e- Government services. Develop a strategic plan to guide e-government services; Understand the needs of all segments of public; and, enable citizens to participate in the design of e-government services; Use well established system development practices to carry out the day-to-day activities of developing, implementing and maintaining e- Government services; Create a learning organization where employees are encouraged to participate in developing and managing e-government services; Develop effective ICT governance mechanisms to assign roles and responsibilities for managing and making decisions about e-government services; ASE 2014 ISBN: 978-1-62561-003-4 3
Develop ICT capabilities focusing on building a suitable ICT infrastructure to sustain long-term investments in e-government; Provide a secure experience for web visitors by developing an e-government security and disaster recovery plan. 3. New QFD Integrated Methodology for Design of E-Government System in Egypt [3] The research study addresses the issue of requirements traceability by assessing the degrees of impact with the help of quality function deployment (QFD), House of Quality (HoQ). HoQ from (QFD) is used to map the voice of customers into multiple phases of Product/Service development life cycle. The advantage of using HoQ (from QFD) is that it traces customer requirements from the very beginning to Product/Service design. As a result, it is easier for both customers and developers to visualize which design item reflects which set of requirements and to what extent these requirements are implemented. Based on the assessment result, limited resources can be allocated to more important design items and the resultant product/service will achieve a higher level of customer satisfaction. Thus, a priority assessment framework, different from the traditional linkage-based traceability methods, is provided to help find the important design items phase by phase. An integrated framework for the application of QFD of E-Government System is developed. The methodology uses the AHP in order to prioritize citizen requirement, the automated HoQ to trace the citizen requirement in multiple phases of design process and TRIZ technique to solve for the contradicting requirements. The HoQ correlates both the Functional requirements and Quality (non-functional) requirements with citizen requirements. In this new integrated methodology a new integrated framework that improves the quality of both the software development process and the product and assures that customers receive high quality products. It is integrated in the sense that it improves the limitations of previous methodologies. It ensures the following: Overcome the vagueness and inconsistencies of customer requirements and elaborate the importance index of those requirements. Utilizing a number of design phases of software development life cycle. Overcome the Limitations due to the difficulty of manual construction and calculations of the huge number of QFD matrices which are time consuming and liable to inaccuracies. Using theory of inventive problem solving (TRIZ) to solve contradictions between technical requirements Analytical Hierarchy Process (AHP) is used for the purpose of prioritizing the customer requirements during implementing the House of Quality. A Questionnaire form (Table 1) is introduced to the customers or users (from different segments) of the system for deciding on the degree of importance of one requirement over another (pair-wise comparisons). The upper half of Table 2 is filled from the information obtained from the questionnaires table. The lower half of prioritization matrix is filled based on the participants' feedback from the upper half of the matrix (Reciprocal values). The data is then averaged over normalized columns to estimate the eigenvector of the matrix, which represents the criterion distribution. To do this, firstly, the sum of the columns in the matrix is computed. Each value in the matrix is then divided by the column sum. The output is the normalized matrix. To determine the score of each requirement, average the row in the normalized matrix by dividing each row sum by the number of requirements Table 1: A Questionnaire form for deciding on the importance index of the Customer Requirements Table 2: Analytical Hierarchy Process for Prioritizing Customer Requirements The ability of AHP to test for consistency is one of the method's greatest strengths. The AHP view of consistency is based on the idea of cardinal transitivity. For example, if Requirement A is considered to be two times more important than Requirement B, and Requirement B is considered to be three times more important than Requirement C, then perfect cardinal consistency would imply that Requirement A be considered six times more important than Requirement C. In this way, if the participants judge Requirement A to be less important than Requirement C, it implies that a judgmental error exists and the prioritization matrix is inconsistent. ASE 2014 ISBN: 978-1-62561-003-4 4
Consistency index/random index (CI/RI) ratio is used to check the consistency of the results. To compute the CI/RI ratio, the following steps have to be followed: 1. Calculate the product of the pair-wise comparison matrix and the vector of scores (Matrix Multiplication) and put the result in the Product column. Make sure that the user data is in decimal form. 2. Calculate the ratios of the score and product values obtained from previous step and put the result in the ratio column. 3. Calculate the CI value. CI=(average ( Ratio Column ) - number of requirements) / number of requirements minus one) 4. Calculate the CI/RI score: To calculate the CI/RI score, the standard RI value is obtained from Saaty's information; a few of those RI values are listed in Table 3. If the CI/RI is smaller than 0.10, then the degree of consistency is satisfactory; however, if the CI/RI is larger than 0.10, inconsistencies exist and the AHP method may not yield meaningful results. Table 3: Random index (RI) values The number of phases of software design process can be three phases or more depending on the application type, the detail of specification required, and the amount of effort and time required. The four-phase approach includes the following types of matrices (as seen in Figure 6). During the design and development phases, it is helpful to know what the most important design items are in terms of their correlation with the requirements. Thus, a priority assessment framework is provided to help find the important design items phase by phase. In this framework, HoQ incorporates customer requirements into multiple phases of the software development life cycle, including system, subsystem, module, and component designs. The advantage of using HoQ (from QFD) in this methodology is that it traces customer requirements from the very beginning to component design. Phase 1: Customer requirements are deployed to both the product functions and the quality factors of the whole system. Phase 2: The product characteristics, which reflect the voice of customers, obtained from the previous phase are deployed into the important subsystem functions and subsystem constraints. Phase 3: The most important functions and constraints of the modules are identified. The subsystem characteristics from the previous phase are deployed to these functions and constraints. Phase 4: Module functions and module constraints are deployed into component functions and component constraints in this phase. In order to provide traceability, the final outputs from each phase are used as the inputs for the next phase matrices. In this manner, customer requirements are incorporated into the whole system. Quality and functionalities are the two major issues affecting the degree of customer satisfaction. Thus, customer requirements are interrelated with both functionalities and quality factors (constraints) using HoQs. The HoQ relating customer requirements with the quality factors is given the name Q-HoQ; similarly, the HoQ relating customer requirements with the functionalities is given the name F-HoQ. The design point analysis matrix is then used to combine the quality factors and functionalities, both of which now have weight values reflecting the impacts from the customer requirements. In Figure 6, the matrices R2, S2, M2 and C2 are Q- HoQs; the matrices R1, S1, M1 and C1 constitute F- HoQs; and the matrices R3, S3 and M3 are of the type of design point analysis matrix. The customer requirements serve as an input into R1 (F-HoQ) and R2 (Q-HoQ) requirement elicitation matrices. The results of these two requirements elicitation matrices serve as inputs for the R3 matrix. Results of the R3 matrix are used to combine the product functions and quality factors into one set of subsystem-level requirements, which are carried over to Phase 2 of the development life cycle where similar steps are taken Figure 6: Phases of QFD Design Process The house of quality has been automated using excel sheets thereby saving effort and time by using automated ASE 2014 ISBN: 978-1-62561-003-4 5
calculations. Besides, it is flexible and extendable. It gives the possibility of adding or deleting more customer or technical requirements to the HoQ matrix without sacrificing the calculations accuracy. The advantages of automating the QFD, House of Quality are: Saving of time and effort in constructing the house of quality. The automated house of quality is reusable i.e. once constructed; it can be used for other applications as well. Ease in modification of content (voices of customers, technical requirements, relationship and correlation ratings) and data. More acceptability of the automated QFD house of quality especially in the services scenario, where the users deem the house of quality as complex. Ease of inserting more number of voices of customers and/ or technical requirements than initially available in the standard format of the automated house of quality. Automatic calculations of scale-up factor, absolute and relative values save a lot of time and effort. Various types of graphs and charts are automatically prepared for aiding in analysis of the house of quality. The QFD house of quality can be made available simultaneously at the computer terminals of various experts for their comments and suggestions. The automated house of quality can be easily sent as an attachment to an email to geographically distant places. Automated house of quality created on MS Excel is free of cost compared to the other QFD software, as MS Office is most commonly used software in any organization even in the developing nations. The method of operation of the new TRIZ for software Matrix is exactly the same as that used in the other Matrices; the user has to first identify what they wish to improve in their system, then what is it that is preventing them from making the improvement. From here it is then necessary to translate the specifics of the conflict pair into a pair from the Matrix parameter list that most closely matches those specifics. From there, the Matrix will reveal the most likely Inventive Principles used by others to successfully challenge that conflict pair. The new Matrix represents the outcome of studying around 40,000 software patents and design solutions. It has been confirmed that the same 40 Principles found in classical TRIZ are applicable to software. Only some of the terminology of the Principle descriptions has to be changed to better suit the software context, but beyond that, they have stayed the same. The parameter list of the new software matrix is shown in figure 7. Figure 7: Twenty-one Parameters of the New Software Matrix 4. Implementing the New QFD Integrated Methodology for Design of E-Government System in Egypt As an example to demonstrate the methodology, we will implement phase 1 of the E-Government system design process. The matrices R1 of (F-HoQ), R2 of (Q-HoQ), and R3 (Design Point Analysis Matrix) will be completed. The customer requirements serve as an input into R1 (F-HoQ) and R2 (Q-HoQ) requirement elicitation matrices. The results of these two requirements elicitation matrices serve as inputs for the R3 matrix. Results of the R3 matrix are used to combine the product functions and quality factors into one set of second phase (subsystem-level requirements) After conducting some preliminary investigation about the G2C sector, the elicited customer requirements (CRs) are as follows: CR1-Fast Response CR2-Accurate and Efficient Service CR3-Service around the clock CR4-Service anywhere CR5-Save Time, Effort and Money CR6-Easy and Understandable GUI CR7-Sharing Decision Making Process 4.1 AHP Process for customer requirements prioritization ASE 2014 ISBN: 978-1-62561-003-4 6
To prioritize citizen requirements, a questionnaire (using the form in table 1) was conducted on 200 people from different sectors of the society The resultant filled questionnaire form after data collection and analysis appears in table 4. Table 4: Filled AHP Questionnaire form 4.2 Design Process: Phase 1 (Requirements Elicitation Phase) F-HoQ: Matrix R1 The System Requirements - Functional Requirements (SRs) developed in phase 1 to satisfy the abovementioned Customer requirements are: SR1-Qualified and Trained employees SR2-Real time On-Line Service SR3-Integrated Centralized service SR4-Encouraging Citizens Participation SR5-Multi-Language GUI SR6-Easy and Clear Forms SR7-Multichannel Citizen Interaction And the F-HoQ Matrix (R1) is shown in figure 8. The AHP matrix is shown in table 5: Table 5: AHP Matrix It appears from the AHP Matrix that the priorities of Customer Requirements Priorities are as follows: Requirement # Priority Index CR2 0.371 CR3 0.262 CR4 0.148 CR1 0.103 CR6 0.055 CR5 0.036 CR7 0.025 Sum 1 Q-HoQ: Matrix R2 Figure 8: F-HoQ Matrix (R1) The Quality Factors Non-Functional Requirements (QFs) developed in phase 1 to satisfy the abovementioned Customer requirements are: QF1-Efficiency and effectiveness QF2-Adaptability to future needs QF3-System security and privacy QF4-Convenience and reliability QF5-Support and Help Service QF6-Customized Citizen Oriented Service And the Q-HoQ Matrix (R2) is shown in figure 9. ASE 2014 ISBN: 978-1-62561-003-4 7
The method for resolving these contradictions is accessible to anybody. There are 40 solutions (or inventive principles) that are generally applicable to all of these problems. The contradiction matrix shows which of the 40 inventive principles to start with when looking for a solution. Figure 11 shows a subset of a complete contradiction matrix. Figure 9: Q-HoQ Matrix (R2) Design Point Analysis Matrix (R3) The Design Point Analysis Matrix (Matrix R3) which correlates the Functional Requirements with the Quality Factors is shown in figure 10. Figure 10: Design Point Analysis Matrix (R3) Results of the R3 matrix are used to combine the product functions and quality factors into one set of second phase (subsystem-level requirements), and so on. 4.3 Application of TRIZ methodology to solve for Contradicting Requirements Innovation practitioners can all benefit from one of the basic concepts of the Theory of Inventive Problem Solving (TRIZ). The most common kind of contradiction is a trade-off. There are trade-offs in all kinds of problems business, personal, product and service development, and operations as shown in Table 6. Figure11: Section of Contradiction Matrix The first step is deciding what is getting better and what is getting worse (or what is preventing things from improving) in the problem. Consider Figure 9 (Q-HoQ Matrix), in which the "customization of service" gets better and "convenience and reliability" gets worse. The parameters that match best are "adaptability and versatility" (improving) and "system complexity" (worsening). The circled cell in Figure 11 shows the numbers of the principles that have been most popular for solving that class of problems, in descending order. Begin with principle 15, which says to make the system more flexible and more dynamic, making it adaptable without making it complex. Perhaps the system could use physical flexibility (such as building in parts that fold up) or maybe it could be "virtually" flexible (like letting the user select how many "help" hints he wants in a program). 5. Conclusion The issue of requirements traceability by assessing the degrees of impact with the help of quality function deployment (QFD), House of Quality (HoQ) is used to map the voice of customers into multiple phases of Product/Service development life cycle. The advantage of using HoQ (from QFD) is that it traces customer requirements from the very beginning to Product/Service design. As a result, it is easier for both customers and developers to visualize which design item reflects which set of requirements and to what extent these requirements are implemented. Based on the assessment result, limited resources can be allocated to more important design items and the resultant product/service will achieve a higher level of customer satisfaction This study addresses a QFD integrated methodology for improvement of the Egyptian E-Government system by using the Analytical Hierarchy Process (AHP) to prioritize citizen requirements, Quality Function Deployment, House of Quality (HoQ) to trace weighted customer requirements (Functional requirements) and quality requirements (Non-functional ASE 2014 ISBN: 978-1-62561-003-4 8
requirements) through all phases of design, and Theory of Inventive Problem Solving (TRIZ) to solve for the contradictions among requirements using the contradiction matrix and the 40 inventive principles. 6. Future Work The future work to be done is adding more quality tools to this integrated model for E-Government System to further improve the performance of the Product/Service design process. The new quality tools to be added are: Tagushi method for design of experiments (Robust design), Design For Six-Sigma (DFSS), Lean design, and Future TRIZ research and innovations. Also the crisp numbers used in correlations will be replaced with Fuzzy membership functions. References [1] Daniel Veit & Jan Huntgeburth (2014), Foundations of Digital Government - Leading and Managing in the Digital Era, Springer-Verlag Berlin Heidelberg. [2] Herbert Kubicek, Ralf Cimander & Hans Jochen Scholl (2011), Organizational Interoperability in E-Government Lessons from 77 European Good-Practice Cases, Springer-Verlag Berlin Heidelberg. [3] Ahmed Attia (2012), Quality Improvement Techniques for Object Oriented Software, LAMBERT Academic Publishing [4] John Terninko, 1997, The QFD, TRIZ and Taguchi Connection: Customer-Driven Robust Innovation, The Ninth Symposium on Quality Function Deployment, June 10, 1997 [5] Saaty, T. L., 1980, The Analytic Hierarchy Process,. New York, NY: McGraw-Hill. [6] Nancy R. Mead,, 2008, Requirements Prioritization Case Study Using AHP, Software Engineering Institute, Carnegie Mellon University. [7] Valeri Souchkov, 2007, Accelerate Innovation with TRIZ, ICG T&C, 2007 valeri@xtriz.com] [8] Herman Hartmann, Ad Vermeulen and Martine van Beers, 2004, Application of TRIZ in Software Development [9] Xiaoqing (Frank) Liu, Yan Sun, Praveen Inuganti, and Chandra Sekhar Veera, University of Missouri-Rolla, Yuji Kyoya, Toshiba Corporation, 2007. A Methodology for Tracing the Requirements in the Object-Oriented Software Design Process Using Quality Function Deployment, SQP VOL. 9, NO. 4/ 2007, ASQ [10] Kanishka Bedi, U21Global, Singapore, 2006, Automating the Quality Function Deployment House of Quality, U21 Global, Graduate School for Global Leaders. [11] Darrell Mann, 2004, TRIZ For Software, Systematic Innovation. [12] Ma Jianhong, Zhang Quan, Wang Yanling, Zhang Wei, 2009 IEEE, Research and Application of the TRIZ Contradiction Matrix in OOD, DOI 10.1109/WCSE.2009.244. [13] Hesham Mahmoud Ibrahim, 2012, E-Government, Proficiental Management Experts Center (PMEC), Cairo, Egypt. ASE 2014 ISBN: 978-1-62561-003-4 9