International Journal of Scientific & Engineering Research, Volume 4, Issue 9, September-2013 51
ISSN 2229-5518
Root Cause Analysis of Variations in
Construction Tasks and Developing Effective
Strategies to Reduce Variations
Mr. V. S. Tiware1, Dr. S. S. Kulkarni2
Abstract— This paper is aimed at knowing the variations and causes of variation occurrence. The main objective of this research is to develop suggestive method to reduce such variation. The delays in planned work and actual work done during construction stage are considered as variations for this research purpose. The variation in construction task can affect the cost of the project to avoid this, it is necessary to improve organizational practices about delay problem in construction industry. The risk assessment matrix is used to know the effect of such causes on performance.
This study helps in knowing the exact causes of delay occurrences with the help of preparing risk assessment matrix. The study is done on the residential sites for this research purpose. The method developed can be useful for minimize the delay occurrence. The implementation of this method results in increase of PPC (percentage planned completed) from 83.3% to 87.5%
As the delay in construction is an issue this research has its own importance.
Index Terms— construction delay, delay analysis, variation, delay causes, risk assessment matrix.
—————————— ——————————
he construction industry is complicated in nature. And all tasks in construction are interdependent. This interde- pendency of tasks causes much complexity because if one task is delayed, then it may affect all succeeding tasks and
results in reducing productivity of the construction, so proper planning is necessary for better performance. The time is al- ways considered as money, saving time at the time of con- struction is nothing but saving money, which is involved within construction. The delays can be observed at the time of variation in planned work and actual work done. So finding root causes of variation and minimization of this variation is important for construction sector. Consider if any task re- quired 100 labour hours and actually it takes 125 labour hours then the variation is of 25 labour hours.
Variation almost always exists in the construction work pro- cess. The term variation is taken for this research as the time difference between actual planned duration and actual executed duration at site. The data collection is carried out through the prepared questionnaire. It will find the main causes among the various individual causes of variation.
The complete removal of variation is not possible in con- struction industry due to its complex nature. So minimization of the variations in construction tasks is necessary. The varia- tion minimizations will definitely result in saving time and money.
This is necessity of construction industry to complete the
————————————————
• Mr. V. S. Tiware1 is currently pursuing masters degree program in M. Tech. Civil, R I T Sakharale, Dist. Sangli, Maharashtra state, India. (Email: tiwareamit9774@gmail.com)
• Dr. S. S. Kulkarni2 is currently Professor, Department of Civil Engineer- ing, Rajarambapu Institute of Technology,Rajaramnagar, Islampur, Dist. Sangli, Maharashtra, India (Email: sushma.kulkarni@ritindia.edu)
project on time or within time and the main challenge is to avoid the delays. If the project is delayed for long time the respective company’s reputation will damage as well as the loss due to delay is also another problem. So an avoidance of occurrence of delay is the best way to save money, reputation, and relationship with clients.
In this research the study is done only for residential sites
and method is developed. But the same method can be
adoptable for all types of constructions like commercial and
industrial type of construction. Variations from planned time are considered as delays, and these variations may be positive or negative. The positive variations means the actual task is completed after planned task and negative variation means the actual task done before the planned task. The positive var- iation is taken as dangerous for project performance because this will affect on successor activities and which can be harm- ful for the project performance. In this research the positive variations are taken into account as they have high influence on project. The delay cannot be completely avoidable with any suitable method, but it can be minimized by proper means. Various factors are affecting on delay occurrence are like hu- man habits, different site conditions, type of work force, na- ture of work, client or contractors interference etc. The re- search on this kind of issue is helpful to know delays and their causes.
The project delay is major problem in construction indus- try. Many projects experience extensive delays and thereby exceed initial time and cost estimates. Due to complex nature the delay can be occurred and the study is necessary to know the root causes of variations and to know how to overcome the delay problems. The construction industry is a major player in
IJSER © 2013 http://www.ijser.org
International Journal of Scientific & Engineering Research, Volume 4, Issue 9, September-2013 52
ISSN 2229-5518
the economy, generating both, employment and wealth.
The following study gives knowledge about the causes of
variation and remedial methods to overcome such variations.
The causes of delay which leads to exceeding the time and cost
of the project should be eliminated or minimized.
The data collection was done through prepared question- naires. The suitable questionnaires are designed to collect ap- propriate data. The data collection is done from three residen- tial projects. The residential sites with well-prepared bar chart are taken for this purpose. The planned data is compared to actual occurred data to know the occurrences of variation or delay in project progress, the bar chart scheduling is used as base line and the delayed time duration. And analysis of vari- ous causes of variation is done; to know the effect of each cause on the productivity of the project.
For data collection purpose the quantitative method of data
collection is adopted. The collected data is further used for analysis purpose. The collected data is used to know the varia- tions in task, as well as for knowing the individual causes of variation. The individual causes are grouped in main cause groups depending on their characteristics and nature of occur- rence on site. The onsite interviews are taken for this purpose. In which the care is taken that the all levels of authorities are involved in this process i.e. from labour to the higher authori- ty of the project. The necessity of involving all levels of author- ities is to consider the perspective of all levels of authorities towards the delay problem. And also it gives the exact causes of the variation in construction tasks.
Data collection process is carried out on three construction
sites from the Kolhapur, Maharashtra, India region. The 15
weeks data is recorded for data collection. And this data is
used for data analysis.
Questionnaire is prepared with, the site information, inter- viewer position and his authority towards work, type of task and factors causing delay for that task. The questionnaire is divided into two main parts. First part is related to general information for both the company and respondent. Second Part includes the list of the identified causes of delay in con- struction project. These causes are classified into nine groups according to the main sources of delay i.e. pervious task, drawing details, Labour, Management, Material, equipment, Work condition, weathering condition, and finance.
For each cause group two questions were asked i.e. what is
frequency of cause occurrence? And what is the degree of se-
verity of this cause on project delay? Both the frequency of
occurrence and severity were categorized on a five-point scale
system. Frequency of occurrence is categorized as follows:
certain, likely, possible, unlikely and rare (from 5 point to 1 point on scale respectively). Similarly, degree of severity was categorized as follows: catastrophic, major, moderate, minor and low (from 5 to l point scale respectively). From which the severity index and the frequency index of the cause occurrence on site has came to know and which can further used in prep-
aration of the risk assessment matrix.
In this the variation survey will be done at three residen- tial construction sites and then root causes of these variations will be found through questionnaires at different level of management. The site selection process is done with following criteria:
• The site should be residential type.
• Scheduled bar chart should be prepared for the work and the work is followed with scheduled bar chart.
• The site should cover maximum activities within data collection period.
The following three sites are selected based on the criteria for data collection purpose in Kolhapur, Maharashtra, India region. The sites are with well-prepared bar chart.
1. Residential project of Amar Patil engineers and con- tractors, Shivaji peth Kolhapur
2. Palm grove project of Bedekar constructions
Devkar panand road, Kolhapur
3. Eden garden housing project of Shree construction,
Bawda road, Kolhapur
The magnitudes of occurred causes are measured with help of following format on sites.
Sr. no. | Activity name | Planned start date | Actual start date occurred | Variation occurred |
The duration of delay is found out through above format. As each task is started the planned start time is noted from planned bar chart and then the actual start time is noted in next column. Through this data the occurrences of the varia- tion is known to us by whom we go for cause finding process with the help of prepared questionnaires. The difference be- tween planned and actual gives the magnitude of variation or delay.
The grouping of causes is done as per category of the cause and nature of occurrence. The five point ranking system is adopted for ranking the nine main causes. The ranking of causes gives us the effectiveness of particular cause on the performance of the project. The nine causes are ranked with the help of mean score method. With risk assessment matrix the causes impacting maximum risk on the project perfor- mance are worked out for suggestive method development.
The five point ranking system is adopted, in which the 25
respondents related to construction industry are selected. And their responses are taken through frequency index forms and analyzed for following methods for ranking i.e. mean score method. The score is given in between 1 to 5 depending on their frequency of occurrence. This is done to know their per- spective about the causes and their effect on the project per- formance.
IJSER © 2013 http://www.ijser.org
International Journal of Scientific & Engineering Research, Volume 4, Issue 9, September-2013 53
ISSN 2229-5518
The individual causes are carried out through prepared questionnaire and interviewing people on the site. These are selected with different level of authorities; the purpose is to take consideration of perspectives of each level of authority. In this the interview are taken from foremen to management lev- el. The following chart shows the grouping of the individual causes into their main groups.
TABLE NO: 1
INDIVIDUAL CAUSES AND THEIR GROUPING
problems are taken in group Labour. The causes related to miss-management are grouped in Management group. The material source and its demand and supply related causes are grouped in Material group. The cause group equipment in- volves causes related to faulty equipment; less availability etc. likewise work condition group involves the availability of work space, Site layout etc. Weather condition group involves causes related to bad weather condition. And finance group involves causes related to finance supply
The cause groups are analyzed by using mean score meth- od of ranking. For this purpose 25 respondents are selected concerning to civil industry. As they have enough knowledge about causes of variation. The respondent has experience of 2-
10 year in construction industry.
The formula used for this method is as below
Mean score = Σ (f * s/N)
Where, f is frequency of the respondent, s is score given by the respondent (i.e. in between 1 for lowest possibility to 5 for highest possibility)
This five point scale is used to calculate the mean score for
each factor, which is then used to determine the relative rank-
ing of each factor. The low mean score assigned low ranks and
high scores allocated as high ranks. The mean score for each
factor is computed by using the above formula. For ranking purpose the frequency of occurrences of the cause are take out through prepared pointing form. The data can be used for mean score method of ranking and also for calculation of fre- quency index.
TABLE NO: 2
RESULTS FOR MEAN SCORE RANKING METHOD
These are the individual causes which are carried out from residential sites survey and literature review of related stud- ies. The 29 individual causes are grouped in 9 main cause groups. The groups are taken for ranking purpose and then as per risk assessment matrix the individual causes of cause group are considered for minimizing the variations from the construction tasks.
The cause group previous task involves the delay due to
previous tasks which is completed before new task. The cause group drawing details involves causes related to incorrect drawings, change in drawings by client and by contractor etc. The method of execution of labour and other labour related
IJSER © 2013 http://www.ijser.org
International Journal of Scientific & Engineering Research, Volume 4, Issue 9, September-2013 54
ISSN 2229-5518
From above mean score method of ranking it can be con- clude that the cause labour has first rank having mean score of
4.00 and in that the 40% respondent give 5 point, 32% re- spondent give 4 point, 16% respondent give 3 point and 12% respondent give 2 point, and no one give 1 point for this cause. Likewise the other ranking is done; the lowest rank is for cause weather condition. It has mean score of 1.64
The results of mean score method are as follows.
TABLE NO: 3 CAUSES AND THEIR RANKS
Weather condition | 1 4% | 0 | 2 8% | 5 20% | 17 68% | 30.40 |
Finance | 0 | 0 | 0 | 3 12% | 22 88% | 22.40 |
TABLE NO: 5 CAUSES AND THEIR SEVERITY INDEXES
The ranking gives us the idea about causes and general per- spective of respondents about the causes.
Severity index shows the severity of the cause occurrences from which it is come to know that, which cause group is more severely occurred or which group is more responsible for delay. In this method the severity index for each cause is calculated with following formula:
Severity index = Σ a (n/N)*(100/5)
Where, a is constant expressing weighting given to each re- sponse, which ranges from 0 for no influence up to 5 for very high; n = frequency of the responses; and N = total number of responses.
TABLE NO: 4 SEVERITY INDEX TABLE
From above results it is clear that the severity index for cause- labour is highest (83.20%) so it is more affecting cause for de- lay occurrence. And the cause- finance has severity index low- est (22.40%) so this cause does not much affect for delay. The following relation is made for severity index and their impact level on the project performance.
TABLE NO: 6 SEVERITY INDEX AND IMPACT LEVEL
So from obtained results about severity the no one cause group in low and minor category, the causes like equipment, material, weather condition, work condition, finance are in moderate category. Drawing detail and management these causes are in major category. And the causes previous task and labour are in catastrophic category.
The frequency index is calculated by using the following for- mula. The frequency of delay occurrence can be known by following formula:
Frequency Index = Σ a (n/N)*(100/5)
Where, a is the constant expressing weighting given to each response (ranges from 1 for rarely up to 5 for certain), n is the frequency of the responses, and N is total number of respons- es.
IJSER © 2013 http://www.ijser.org
International Journal of Scientific & Engineering Research, Volume 4, Issue 9, September-2013 55
ISSN 2229-5518
TABLE NO: 7 FREQUENCY INDEX TABLE
Weather condition, finance, and work condition are in possi- ble category; and the causes like material, equipment, man- agement are in likely category. And in certain category the causes like previous work, drawing details, and labour are obtained.
The risk assessment matrix is nothing but the matrix show- ing the graph of severity and frequency. For this research pur- pose the 5x5 matrix is used. The severity is grouped in five categorized as low (0-09%), minor (10-19%), Moderate (20-
49%), Major (50-79%), Catastrophic (80-100%) etc. And the frequency is categorized as rare (0-09%), unlikely (10-19%), possible (20-49%), likely (50-74%), certain (75-100%) etc.
DIA: 1 RISK ASSESSMENT MATRIX
TABLE NO: 8 CAUSES AND THEIR FREQUENCY INDEX
L= low risk, M= Moderate risk, H= High risk, E= Extreme risk
The matrix of frequency and severity is formed from obtained results. And the causes are placed in their respective fields. The final output of the result is as follows;
TABLE NO 11: RESULTS OF RISK ASSESSMENT MATRIX
From above results it is clear that the Frequency Index for cause labour is highest (80.00%) so it is frequent cause for de- lay occurrence. And the cause weather condition has Frequen- cy Index lowest (32.80%) so this cause rarely occurred. The following relation is made for Frequency Index and their im- pact level on the project performance.
TABLE NO: 10 FREQUENCY INDEX AND IMPACT LEVEL
Range (%) | Impact level |
0–09 | Rare |
10–19 | Unlikely |
20–49 | Possible |
50–74 | Likely |
75–100 | Certain |
From above chart and obtained results it can be conclude
that the no one cause group is in rare and unlikely category.
IJSER © 2013 http://www.ijser.org
International Journal of Scientific & Engineering Research, Volume 4, Issue 9, September-2013 56
ISSN 2229-5518
TABLE NO: 12 CAUSES AND THEIR RISK ON PROJECT
Cause group | Risk on project |
Labour, Previous task, Drawing details, Management | Extreme |
Equipment, Material, Work condition | High |
Weather condition, Finance | Moderate |
The risk assessment matrix gives idea about the risk pro- duced by the cause group on the project. The causes Labour, Previous task, Drawing details, Management produce extreme risk on project. The causes Equipment, Material, Work condi- tion produces high risk. And the causes Weather condition, Finance produce moderate risk on project performance.
The following process can be adoptable for delay analysis and as preventive measure for delay occurrence. Few steps of this are as mentioned below.
The weekly or monthly survey of scheduled progress and actual progress should be done, through which the variation from base line can be come to know. This will help us to know where the delay in construction tasks are occurred or not, and then we can also find out in which activity delay occurred with its magnitude. The project bar chart can be taken as base line for this survey purpose. And the mapping of project pro- gress can be done.
The feedback system should be adopted to know the cause of delay occurrence. The feedback system should include dif- ferent levels of authorities to know their perspectives about the delay, and to know the exact causes of delay. Also the se- verity and the frequency of such causes occurrence should be taken out from feedback system; from which we can come to know the impact of that cause.
The risk assessment matrix should be formed to know the how much risk is having by that causes of variation. For that purpose the severity data and frequency data can be used and the severity index and frequency index for that causes can be computed with formulae.
And then the causes causing maximum risk can be taken
out for the elimination with proper solution. The risk assess-
ment matrix gives the idea about which causes producing low or high risk on project performance. The another advantage of this is we can concentrate on high degree risk causes than the low degree risk causes which will definitely saves the time of process improvement.
The appropriate solution for the obtained cause of delay should be carried out and the preventive measure should be taken for that cause. The solution should have some technical bases. Some of the suggestive methods are given below.
The proper grouping of work force throughout the organization is necessary; if any task needs 25 labours and there is provision of 20 or 22 labours then there is probability of delay occurrence. And for that the proper distribution of work force of relative tasks should be adopted. The exact allo- cation of suitable work force on different types of work is needed.
The training program should be provided as per the requirement of improvement required. The training should include the solutions about current problems facing by the project progress, i.e. if in any case delay occurred due to lack of communication then there should be training program for improving the communication and coordination.
For drawing details and drawing changes the proper communication system should be adopted in case if commu- nication of drawing details is the cause of the delay. The com- puterized system can be adopted for fast communication of drawing, drawing details and changes. So misguidance about drawing cannot be adopted. Now days mailing the document are very common, cheap and fastest source of communication. If in any case there is less communication with clients then there should be one authority for handling the client side communications. So it will reduce time for decision by client.
The schedule of the project should be revised at some time interval with neglecting previous delays. So the new goal can be formed within the organization.
Percentage planned complete (PPC) to indicate work plan variability. The PPC is the ratio of the number of tasks 100% completed to the number of tasks planned.
The project progress can be analyzed with checking the ra- tio of completed tasks to the planned task in some time inter- val say x i.e. the formula can be as follows:
PPC= no. of tasks completed in x duration / planned task for x duration
If there were 10 tasks planned on a monthly work plan, and at the end of the month, six of those tasks had been completed as per planned; then the PPC would be 60%.The value of PPC should be near to one; then only it can be said that the project is as per scheduled time. Lower PPC is understood to indicate greater work flow variation.
IJSER © 2013 http://www.ijser.org
International Journal of Scientific & Engineering Research, Volume 4, Issue 9, September-2013 57
ISSN 2229-5518
The main thing is to repeat the process for continuous im- provement after definite time interval i.e. after one or three month for better results and fast improvement in delay reduc- tion. The repeating the process gives the exact causes of delay and can be minimized at the time of occurrence.
The above method is implemented on residential site to check the feasibility of this method. The following results are observed. For feasibility analysis PPC ratio before implemen- tation of method and PPC ratio after implementation is worked out.
PPC = no. of tasks completed in one month / planned task for one month
=5/6
= 0.833
PPC = no. of tasks completed in one month / planned task for one month
=6/7
=0.875
So we can conclude that the previous PPC ratio is 0.833 and the current PPC ratio is 0.875.
The previous PPC ratio is less than the current PPC ratio, hence the project performance is improved and which is bene- ficial.
In this the nine cause groups are taken for ranking purpose. The cause labour has 1st rank; it has mean score of 4.00 out of
5. The cause drawing details has mean score of 3.96 with 2nd rank. Simultaneously the causes like previous tasks, manage- ment, equipment, material, work condition, finance, and weather condition are ranked form 3rd rank to 9th rank. The ranking of causes concludes that the impact of the respective causes from the perspective of respondent.
The causes like Equipment, Material, Weather condition, Work condition, Finance has moderate severity on project progress as they have severity index ranges from 20-49%. And the Drawing detail and management these causes has major severity with severity index in between 50-79%. And the caus- es previous task and labour has catastrophic severity between
80-100%.
The causes Weather condition, finance, and work condition
has possible frequencies as they have frequency index in be-
tween 20-49%; and the causes like material, equipment, man-
agement are likely occurred with frequency index ranges in
50-74%. And in certainly occurring causes are previous work,
drawing details, and labour with frequency index in between
75-100%.
Risk assessment matrix concluded that the causes like la- bour, previous tasks, drawing details and management pro- duces the maximum risk on the project progress. And these are the main causes of variations in construction tasks.
The causes like material, work condition, equipment pro-
duce high risk on project progress. And the causes like weath-
er condition and finance produce moderate risk on project
progress.
The implemented method result shows the PPC ratio before implementing is 0.833 and PPC ratio after implementing de- veloped method is 0.875. It means the PPC ratio is increased by adopting this method. The implementation of this method results in increased PPC from 83.3% to 87.5% and from this it can be concluded that the project performance is increased.
The study is done on the residential type of project. But the same study can be adoptable on the other type of pro- jects like commercial and industrial sites. The complications of this study such as interviewing and calculations can be mini- mized with making appropriate software for delay analysis. This will give the fast results about the delay. The risk assess- ment matrix can be used for other constraints in civil engineer- ing like waste minimization, labour productivity and its effect on performance etc.
[1] Wambeke, Brad William(2011) “Identifying, Prioritizing, and
Reducing Variation of Construction Related Tasks”
[2] Wambeke, B., Hsiang, S., and Liu, M. (2011). “Causes of Variation
in Construction Project Task Starting Times and Duration.” J.
Constr. Eng. Manage., 137(9), 663–677
[3] Wambeke, B., Liu, M., and Hsiang, S. (2012). “Using Last Planner
and a Risk Assessment Matrix to Reduce Variation in Mechanical
Related Construction Tasks.” J. Constr. Eng. Manage., 138(4), 491–
498.
[4] Howell, G. A., and Ballard, G. (1994). “Implementing lean con-
struction: Reducing inflow variation.” 2nd Annual Conference of
the international group for Lean Construction, Catolica Univer- sidad de Chile, Santiago, Chile.
[5] Min Liu, A.M.ASCE; Glenn Ballard, M.ASCE; and William Ibbs, M.ASCE (2011) “Work Flow Variation and Labor Productivity: Case Study” Journal of Management in Engineering, Vol27,No.4,October2011,pp236-242
(doi http://dx.doi.org/10.1061/(ASCE)ME.1943-5479.0000056)
[6] Mohammad, A.I. Che Ani1, R.A.O.K. Rakmat, M. A. Yusof (2010) “Investigation on the causes of variation orders in the construction of building project – a study in the state of Selangor, Malaysia”
IJSER © 2013 http://www.ijser.org
International Journal of Scientific & Engineering Research, Volume 4, Issue 9, September-2013 58
ISSN 2229-5518
Journal of Building Performance
[7] MuraliSambasivan *, Yau Wen Soon(2006) “Causes and effects of
delays in Malaysian construction industry” International Journal
of Project Management25(2007)517–526
[8] Report on “Construction Delays in Florida: An Empirical Study”
by Syed M. Ahmed, Ph.D., M.ASCE, Salman Azhar, M. Engg., M.ASCE, Mr. Mauricio Castillo Ms. PragnyaKappagantula
[9] Sadi A. Assaf , Sadiq Al-Hejji(2005)“Causes of delay in large con-
struction projects” International Journal of Project Management 24 (2006) 349–357
[10] Ibrahim Mahamid; AmundBruland; and Nabil Dmaidi (2012) “Causes of Delay in Road Construction Projects” Journal Of Man- agement in Engineering © ASCE / July 2012
[11] Azlan Shah Ali, Andrew Smith, Michael Pitt and Chan Hong
Choon “Contractors’ perception of factors contributing to project
delay: case studies of commercial projects in Klang valley, Malay- sia”
[12] Alwi, Sugiharto and Hampson, Keith (2003) “Identifying the im- portant causes of delays in building construction projects”. The
9th East Asia-Pacific Conference on Structural Engineering and Construction, Bali, Indonesia. Accessed from: https://eprints.qut.edu.au/secure/00004156/01/Bali_Conference
IJSER © 2013 http://www.ijser.org