Scholarly article on topic 'Quality Risk Management Model for Railway Construction Projects'

Quality Risk Management Model for Railway Construction Projects Academic research paper on "Agriculture, forestry, and fisheries"

CC BY-NC-ND
0
0
Share paper
Academic journal
Procedia Engineering
OECD Field of science
Keywords
{"railway construction project" / "risk management" / "a figure and four tables method" / MIS}

Abstract of research paper on Agriculture, forestry, and fisheries, author of scientific article — Li Qing, Liu Rengkui, Zhang Jun, Sun Quanxin

Abstract Information technology has significant potential to enhance the engineering quality of the risk management of railway construction projects. The Shanghai Railway Bureau has promoted a risk management method based on the “A Figure and Four Tables” method (AFFTM) to assess the engineering quality; however, this method still suffers from several problems in railway construction project management. In this paper, we have combined the concepts and processes of the AFFTM with those of information technology and presented the implementation scheme of a new risk management system—the railway construction project quality risk management information system (RCPQRMIS)—that can be used to design and develop workable information tools for quality risk management. The paper analyzes the data standards of RCPQRMIS and creates a model for dynamically tracking the quality risk (“quality risk dynamic tracking” model) for providing pre-warning information on quality risk (“quality risk pre-warning” model) and for automatically generating publicity parameters for quality risk (“automatically generated quality risk publicity figure” model). The proposed system enables the visualization of the quality associated with the risk control, dynamic tracking, automatic pre-warning, and closed-loop management of railway construction projects. In addition, this paper presents the functional modules of the RCPQRMIS and its practical applications. Our application results show that the system successfully realized unified management of risk source information and multi-level sharing. In this manner, by using our system, we were able to significantly improve real-time tracking and pre-warning of the risk state, automatic generation of quality risk publicity figures, efficiency, and risk management levels.

Academic research paper on topic "Quality Risk Management Model for Railway Construction Projects"

Available online at www.sciencedirect.com

ScienceDirect Procedía

Engineering

Procedía Engineering 84 (2014) 195 - 203 ;

www.elsevier.com/locate/proeedia

"2014 ISSST", 2014 International Symposium on Safety Science and Technology

Quality risk management model for railway construction projects

LI Qinga, LIU Rengkuia *, ZHANG Junb, SUN Quanxinc

aState Key Laboratory of Rail Traffic Control and Safety, Beijing Jiaotong University, Beijing 100044, China bShanghai Railway Bureau, Construction Management Department, Shanghai 200071, China cMOE Key Laboratory for Urban Transportation Complex Systems Theory and Technology, Beijing Jiaotong University, Beijing 100044, China

Abstract

Information technology has significant potential to enhance the engineering quality of the risk management of railway construction projects. The Shanghai Railway Bureau has promoted a risk management method based on the "A Figure and Four Tables" method (AFFTM) to assess the engineering quality; however, this method still suffers from several problems in railway construction project management. In this paper, we have combined the concepts and processes of the AFFTM with those of information technology and presented the implementation scheme of a new risk management system—the railway construction project quality risk management information system (RCPQRMIS)—that can be used to design and develop workable information tools for quality risk management. The paper analyzes the data standards of RCPQRMIS and creates a model for dynamically tracking the quality risk ("quality risk dynamic tracking" model) for providing pre-warning information on quality risk ("quality risk pre-warning" model) and for automatically generating publicity parameters for quality risk ("automatically generated quality risk publicity figure" model). The proposed system enables the visualization of the quality associated with the risk control, dynamic tracking, automatic pre-warning, and closed-loop management of railway construction projects. In addition, this paper presents the functional modules of the RCPQRMIS and its practical applications. Our application results show that the system successfully realized unified management of risk source information and multi-level sharing. In this manner, by using our system, we were able to significantly improve real-time tracking and pre-warning of the risk state, automatic generation of quality risk publicity figures, efficiency, and risk management levels.

© 2014 The Authors. Published by ElsevierLtd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).

Peer-review under responsibility of scientific committee of Beijing Institute of Technology Keywords: railway construction project; risk management; a figure and four tables method; MIS

CrossMar]

* Corresponding author. Tel.: +86-139-1068-8169; fax: +010-51688554. E-mail address: rkliu@bjtu.edu.cn

1877-7058 © 2014 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).

Peer-review under responsibility of scientific committee of Beijing Institute of Technology doi: 10.1016/j .proeng.2014.10.426

1. Introduction

Engineering products delivered by construction enterprises to owners have various quality attributes. Deviations from the expected values of these attributes can lead to loss of benefits and uncertainties for both the owner and the construction enterprise [1]. Such deviations are typically referred to as the quality risks of the engineering project.

Railway projects are characterized by large-scale operations, modern technology, complex structures, high technical and quality standards, long durations, and collaborative units. These factors make it necessary to perform railway construction project management in consideration of the risks involved. If the risks are not prevented effectively, they may hinder the smooth realization of the construction goals, which may even lead to serious consequences [2,3]. While railway authorities have long attached great importance to risk management, there remain many problems and hidden dangers, mainly caused by irregularities, delays, and lack of management, that affect safety. These problems are mainly caused by unclear duties and responsibilities of each department, lack of risk control measures, untimely risk treatments, etc. [4]. Given the problems existing in the risk management of railway engineering construction quality, it is imperative for builders to employ risk management theories and methods to constantly improve and perfect the quality management systems used for railway engineering and technology.

The Shanghai Railway Bureau has proposed a risk management system for the construction engineering quality of railway projects based on the "A Figure and Four Tables" method (AFFTM) used in railway construction project management practice [5]. The Shanghai Railway Bureau has successfully applied this system to the management of railway construction projects with remarkable results.

The AFFTM involves the application of risk management theory on quality to integrate risk identification, analysis, assessment, treatment, tracking, and post-assessment. In this method, the quality risk publicity figure, quality risk analysis and identification table, quality risk treatment responsibility table, quality risk dynamic tracking table, and quality risk treatment evaluation table are used to achieve improved and comprehensive system optimization [5]. This method smoothens and standardizes the risk management procedures for the construction engineering quality of railway projects. Moreover, risk management becomes more standardized, clear, scientific, accurate, and effective by avoiding and controlling project quality risks. Hence, risk management such as that based on the AFFTM method is extremely useful in railway construction and management [6].

However, in practice, the AFFTM still suffers from several disadvantages in railway construction project management—(1) the production cycle for quality risk publicity figures is long; (2) four tables have to be manually filled, which is tedious and time-consuming; (3) there is a lack of standardization; (4) information sharing and comprehensive utilization of data are difficult; (5) the variation in risk state cannot be analyzed in detail because these problems are constrained its large-scale promotion and scope of application; and (6) difficulties are faced in providing real-time status information of risk pre-warning to the responsible people.

To solve these problems, in this paper, we have proposed a railway construction project quality risk management information system (RCPQRMIS) that is based on the AFFTM. We discuss the data standards of the RCPQRMIS; models for dynamically tracking quality risks, pre-warning quality risks, and automatically generating publicity figures for quality risks; and the functional modules of the RCPQRMIS and its practical applications.

2. Data analysis model for RCPQRMIS

2.1. Source and standardization of data

Four types of quality risk control tables—quality risk analysis and identification table, quality risk treatment responsibility table, quality risk dynamic tracking table, and quality risk treatment evaluation table—and one figure—quality risk publicity figure—were used as the main data sources for building the RCPQRMIS. The standardization of the main data source, including the standardization of the data content, coding system, and update frequency, served as the foundation of computer management. We designed the standardization based on the actual requirements of the AFFTM and the general principles for building computer databases. Fig. 1 shows the standardized structure of the data content of the AFFTM, and Table 1 lists the input units and update frequency requirements of the main data source of the AFFTM.

Fig. 1. Main data standard of RCPQRMIS. Table 1. Input units and update frequency requirements of main data source.

Data source

Content

Input units

Update frequency

Quality risk analysis and identification table

Quality risk treatment responsibility table

Quality risk dynamic tracking table

Quality risk treatment evaluation table

Quality risk publicity figure

Table formed by records of analysis and identification of the risk sources from headquarters by organizing exploration, design, supervision, and construction units.

Table formed by records of the risk treatment measures developed and assigned to the responsible departments and persons by headquarters according to risk sources.

Table formed by records of risk treatment measures and evaluations of the effects of risk treatment measures and residual risks.

Table formed by records of comprehensive evaluations of the final status of all quality risks after taking risk treatment measures.

Thematic map formed by risk-related information marked in the entire planar graph according to the risk events, factors, and levels.

Headquarters

Construction units

Construction units

Headquarters

Headquarters

Before the start of project and finding new risk sources

After recording the data of quality risk analysis and identification table

At the end of the daily construction activity

After finishing the treatment of the risk

When changes in risk resource database are updated

Historical experience regarding the design units, daily quality monitoring, and owner management mode for the risk source were mainly obtained from the design units, daily quality monitoring, and historical management experience of the owners. Among them, the Ministry of Railways had formulated strict quality control standards and accelerated systems. Daily quality monitoring included regular grading quality checks and quality inspections. In the quality control process, quality problems were graded by severity; serious quality problems were raised to the risk source and recorded in the quality risk analysis and identification table.

2.2. Quality risk dynamic tracking data analysis model of railway construction project

For a particular risk source, the data of the quality risk control table shown in Fig. 1 could constitute a time series. Although some relationships existed between the data, artificial methods based on AFFTM could not easily determine the same in practice. Therefore, the authors utilized a time series analysis technique to develop the quality risk dynamic tracking data analysis model for a railway construction project (hereafter referred to as "quality risk dynamic tracking model"). The relationships were automatically calculated using a computer, and they could help managers to analyze the existing problems in the historical management of a risk source as well as the future trend of the changes. This model is the first one that can be used to automatically analyze the trajectory and characteristics of risk source state changes in railway construction engineering.

This model was implemented as follows. First, a set of key performance indicators (KPIs) of the risk source states was developed based on the data from the quality risk control tables [7]. The trajectory curve of the KPIs was then plotted using a rectangular coordinate system, where the X-axis represented time and the Y-axis, a particular KPI. Finally, the characteristics of each KPI curve were analyzed [8]. In the model, the KPI can be selected based on the requirements of risk state management; however, it must be calculated according to the historical data recorded in the quality risk control tables. For example, the set of KPIs can contain risk probability, risk degree, rate of variations in risk degree, and residual risk level.

The KPI of the quality risk can be set as Q., where j e {1,2,- • -,m} and m represents the total number of indicators

in the KPI system. ti and Kt represent the indicator values of Q , where /e {1,2,- . vy represents the time

K — K

variation rate of Ki : vi = —. Fig. 2 shows the historical variation trajectory curves of Qj ; the X-axis

represents time and the Y-axis, Qj . Here, Qo , Qm , Qa , Qb , and Qc are the pre-warning thresholds of Qj ; their

specific roles are described in detail in Section 2.3. The variation characteristics of the trajectory curves can be analysed using the following method.

Fig. 2. Schematic diagram of q curves.

The proposed model was first used to calculate the mean values K and mean square errors SK of the time sequence data K = [Kx, K2 >..., k,..., Kn ] of Q as follows:

Here, K represents the average risk level of quality risks and 5K , the entire historical fluctuation conditions of the quality risks. If 5K is small, historically, the changes in the quality risks are stable. On the other hand, if 5K is large, the changes in the quality risks show leaps, indicating that some problems may have been encountered in risk control management by the managers.

The proposed model was then used to calculate the mean values v and mean square errors Sv of the time sequence data vi = {v2,v3,...,vn} as follows:

Here, v represents the average rate of changes in the quality risks and Sv, the entire fluctuation conditions of the time changing rate of quality risks. If Sv is small, the rate of changes of the quality risks' time changing rate is stable. Large Sv indicates that the quality risks' changes have a large acceleration historically; hence, managers should focus on these risks because the probability of significant changes in them is greater.

In this manner, this model attempts to capture the entire fluctuation tendencies of the quality risk indicator Qj

historically and analyze the rules of risk variations.

2.3. Quality risk pre-warning data analysis model for railway construction project

Timely perception of the risk status and alerts about a rapidly deteriorating risk state are critical steps for risk control, and they remain difficult problems requiring urgent solutions in the field of railway construction. Based on the model proposed in Section 2.2, through classification management for the risk source state index of KPIs, the authors proposed a quality risk pre-warning data analysis model for railway construction projects (hereafter referred to as "quality risk pre-warning model"). This model used a computer for automatic judgment, automatic association of responsible people, and automatically sending pre-warning information, and it is the first to realize automatic risk pre-warning in railway construction engineering. Our implementation method [9] provides managers with automatic quality risk pre-warning information across different risk levels that are set according to a threshold range by setting different warning levels, Qj, based on the changes in Qj provided by the quality risk dynamic tracking model. In

Fig. 2, Qo is the planned prewired safety production value; Qm , the achieved optimal safety production value during

actual production; Qa, the prepared pre-warning value; Qb, the threshold pre-warning value; and Qc, the danger pre-

warning value. Thus, [Qa,Qm] represents the normal production regions; [Qb,Qa], the pre-warning prepared regions;

[Qc,Qb], the pre-warning regions; and [0, Qc], the danger pre-warning regions. Az represents the pre-warning type,

where z e {0,1,2,3} .

The model is shown in equation (5) as follows:

A0 (No pre-warning message) K e(Ka, Km ]

A1 (Send prepared pre-warning message) K Kb, Ka ]

A2 (Sendpre-warning message) K e(Kc,Kb ]

A3 (Send danger pre-warning message) K e(ü,Kc ]

By using the pre-warning model, once the quality risk control table is updated, the software will automatically determine the warning level, make judgments regarding who should receive messages, and send text messages automatically using an appropriate messaging platform.

2.4. Data model for automatically generated quality risk publicity figure for railway construction project

It is essential to visualize risk information to improve the utilization efficiency of risk data. In practice, the quality risk publicity figure of AFFTM cannot easily be rendered artificially and cannot be updated in a timely manner. The authors used the mileage attribute in quality risk control tables to establish a conversion algorithm for converting mileage coordinates to latitude and longitude coordinates and to develop a model for automatically generating the quality risk publicity figure. This figure was automatically drawn using a computer, and it is the first example of automatic risk information visualization in railway construction engineering.

The software takes advantage of the data model for the automatically generated quality risk publicity figure (hereafter referred to as "automatically generated quality risk publicity figure model") to automatically tag information about risk events, risk positions, and risk levels that is provided by the quality risk control table to the quality risk publicity figure. The basic map of the quality risk publicity figure was obtained from the design units; it typically uses two-dimensional latitude and longitude coordinates. However, the coordinates of the risk source in the quality risk control table were one-dimensional distance coordinates of railway lines. Therefore, the key to determining the automatically generated quality risk publicity figure was a conversion algorithm to convert the one-dimensional mileage coordinates to two-dimensional latitude and longitude coordinates.

The algorithm for the automatically generated quality risk publicity figure model is given as follows:

Step 1. Data preparation: The entire line plane figure and line control point database of the latitude and longitude coordinates are obtained from the design units.

Step 2. Data source of distance coordinates of risk source: obtained from the one-dimensional mileage coordinate L of some risk R using distance information from the quality risk analysis and identification table.

Step 3. Data source of control point coordinates of risk source: obtained from the line control points R1 and R2 found in the database adjacent to the risk source R; the mileage coordinate L1 and latitude and longitude coordinates (x1,y1) of R1 as well as the mileage coordinate L2 and latitude and longitude coordinates (x2,y2) of R2 are obtained from the control point database.

Step 4. Calculation of latitude and longitude coordinates of risk source: calculating the latitude and longitude coordinates of R using equation (6) [10] and the distance coordinates of the risk source R and the latitude and longitude coordinates of the adjacent control points R1 and R2.

Step 5. Data source of labelling information of risk source: obtained from the risk events, risk levels, and risk positions of risk source R from the quality risk control table.

Step 6. Labelling the risk source information: realizing automatic labelling on the line plane figure and layout of the information text boxes of the risk source R.

By using this algorithm, the software can dynamically generate the quality risk publicity figure, thereby ensuring that the information on the quality risk analysis and identification table is synchronous with that of the quality risk publicity figure. Moreover, in comparison with the typical artificial generation method, the efficiency is improved while saving manpower and material resources.

y = y1 +

x = x1 +

( x2 - x1)(L - L1)

L2 - L1 ( y 2 - y1)(L - L1) L2 - L1

3. Function design of RCPQRMIS

3.1. Overview of system functions

The system contains seven main function modules—quality risk publicity figure function, quality risk identification and analysis function, quality risk treatment responsibility function, quality risk dynamic tracking function, quality risk treatment evaluation function, quality risk pre-warning function, and system maintenance and management—as shown in Fig. 3.

3.2. Function design

The quality risk publicity figure function automatically and dynamically generates the model using the quality risk publicity figure of railways on the basis of the recognized quality risks and thereby realizes a spatial dynamic visualization expression for quality risk information.

The high-speed railway construction project quality risk management information system

Quality risk analysis and identification

Quality risk analysis and identification table record

Quality risk analysis and identification table query

Quality risk analysis and identification table statistics

Quality risk publicity figure

Quality risk

publicity figure check

Quality risk to

view detailed information

Quality risk dynamic tracking

The quality risk

warning ; , i ~

Analysis of quality risk warning

Quality risk

warning information query

Quality risk treatment evaluation

Quality risk treatment responsibility

Quality risk

warning information statistics

Quality risk treatment responsibility table record

Quality risk treatment responsibility table query

Quality risk treatment responsibility table statistics

System maintenance management

Quality risk Quality risk Quality risk Quality risk Quality risk Quality risk Quality risk Post

dynamic dynamic dynamic dynamic treatment treatment treatment Organization User Rights System log

tracking table tracking table tracking table tracking table evaluation evaluation evaluation management management management management management

record query statistics check table record table query table statistics

Fig. 3. Functions of system.

The quality risk identification and analysis function automatically assigns a unique code to the risk sources on the basis of the quality risk information filled by users, thereby realizing convenient queries on and statistical analyses of risk source information.

The quality risk treatment responsibility function helps users to prepare a risk response plan, create preventive measures for each risk, and ensure that all control measures are allocated to the responsible departments and persons. In this manner, convenient queries and statistical analyses can be realized for the quality risk treatment responsibility information based on the data from the quality risk analysis and identification table.

The quality risk dynamic tracking function can help users to record measures while dealing with risk events and the state of the risk source according to the quality risk treatment responsibility table, thereby realizing convenient queries on and statistical analyses of the risk dynamic tracking information. The variation trends of a risk can be analyzed through the variation graphs of the risk source Qj using the quality risk dynamic tracking model. In this

manner, managers can understand the variation characteristics of each risk source state with time and accumulate data for the quality risk pre-warning function.

The quality risk pre-warning function achieves effective control via messages announced by an intelligent mobile terminal to the relevant responsible persons according to the scope of the threshold where the risk source trajectory Qj is located, thereby improving the alertness of the relevant responsible persons for risks and achieving

effective risk control.

The quality risk treatment evaluation function helps users to clearly record the details of the risk treatment process and provides comprehensive evaluation results of the final state of the risk after the risk treatment. It also records residual risks that cannot be prevented completely and creates follow-up disposal measures according to the quality risk dynamic tracking table. In this manner, this function realizes convenient queries on and statistical

analyses of the risk treatment evaluation information, closed-loop management of risks, and the basis for risk observation and monitoring during operation.

4. Analysis of practical applications

The distribution scope of the system's users is wide and involves the construction headquarters, construction units, and supervision units. To realize real-time data sharing, the system provides comprehensive interactive realtime information for the risk management of the quality of railway construction projects using the Internet and VPN; Oracle database technology and a B/S architecture are used for this purpose. Fig. 4 shows a risk publicity figure that was automatically generated by the system using actual data. After a pilot application was implemented in the Shanghai Railway Bureau, the following significant results were achieved:

1) The system provided an information tool for AFFTM that is convenient for large-scale promotion and application.

2) The system could be used to visualize railway construction project quality risk control and improve the efficiency of quality risk control.

3) The system could help in preparing a quality risk treatment responsibility table, making the risk treatment process more focused and scientifically rational.

4) The system could help in achieving dynamic tracking of the railway construction project quality risk treatment process, from the original summary assessment once a week to real-time tracking using a computer, allowing managers to accurately grasp the trend of each risk and making the quality of risk treatment more timely and efficient.

5) The system could change the traditionally released early-warning information one time a week, realize the real-time evaluation and automatic transmission of quality risk pre-warning information, and achieve the objectives of timely prevention and pre-control of quality risks.

6) The system could help in evaluating quality risk treatment results to achieve closed-loop quality risk management.

Fig. 4. Risk publicity figure.

5. Conclusions

In this paper, we have discussed the development of the RCPQRMIS in detail by first analyzing the data standards and subsequently establishing a quality risk dynamic tracking model, quality risk pre-warning model, and automatically generated quality risk publicity figure model. Finally, we summarized the effects of the application of the management information system. Our system successfully realized the visualization of the quality of risk control of railway construction projects, dynamic tracking, automatic pre-warning, closed-loop management, and other features that tremendously improve railway construction project risk management. In future studies, we will investigate methods to improve the data quality and to achieve further integration of the system functions in the risk management of railway construction projects.

Acknowledgements

This research was funded by the China Railways Technology R&D Program under Grant 2013G009-I. References

[1] Fisk E R. Construction project administration[M]. 7th ed. Pearson Education, 2003.

[2] Yongfu Sun. The full implementation of modern project management in railway construction[M]. Journal of railway construction project management. Beijing: China Railway Publishing House.2004.209-217.

[3] Feng Wang. The high-speed railway construction project quality risk management theory and practice[M]. Railway engineering construction management of exploration and practice. Shanghai Railway Bureau. Beijing: Tsinghua University press. 2013.

[4] Guangzu Sheng. Implementation of the safety risk management to achieve development of railway safety[M]. Safety risk management talk. Mor Political Department Propaganda Department. Beijing: China Railway Publishing House.2013. 2-5.

[5] Feng Wang. High speed railway construction engineering quality risk management "a figure four table method"[M]. Railway engineering construction management of exploration and practice. Shanghai Railway Bureau. Beijing: Tsinghua University press.2013. 2-12.

[6] Jianjia Sun. Application of risk management "a figure four table method" in Anqing in the Yangtze River bridge project [M]. Railway engineering construction management of exploration and practice. Shanghai Railway Bureau. Beijing: Tsinghua University press.2013.49-56.

[7] Duncan W R. A guide to the project management body of knowledge (PMBOK Guide)[M]. 4th ed. USA: Project Management Institute, 2008.

[8] Huashan Sun. Risk management of production safety[M]. Beijing: Chemical Industry Press. 2006.

[9] Shusheng Zhao. Longxiang Zhang. Railway construction project management[M].Beijing. China Railway Publishing House. 2004: 170-173.

[10] Zhangwei Ji. Study on railway construction project management integrated data platform [D]. Beijing: Beijing Jiaotong University. 2007.