Decision Tree Approach to Classify and Quantify Cumulative Impact of Change Orders on Productivity

Multiple or unusual change orders often cause productivity losses through a “ripple effect” or “cumulative impact” of changes. Many courts and administrative boards recognize that there is cumulative impact above and beyond the change itself. However, determination of the impact and its cost is difficult due to the interconnected nature of construction work and the difficulty in isolating causal factors and their effects. As a result, it is very difficult for owners and contractors to agree on equitable adjustments for the cumulative impact. What is needed is a reliable method (model) to identify and quantify the loss of productivity caused by the cumulative impact of change orders. A number of studies have attempted to quantify the impact of change orders on the project costs and schedule. Many of these attempted to develop regression models to quantify the loss. However, traditional regression analysis has shortcomings in dealing with highly correlated multivariable data. Moreover, regression analysis has shown limited success when dealing with many qualitative or noisy input factors. Classification and regression tree methods have the ability to deal with these complex multifactor modeling problems. This study develops decision tree models to classify and quantify the labor productivity losses that are caused by the cumulative impact of change orders for electrical and mechanical projects. The results show that decision tree models give significantly improved results for classification and quantification compared to traditional statistical methods in the field of construction productivity data analysis, which is characterized by noisiness and uncertainty.     

Impact of Change Orders on Labor Efficiency for Electrical Construction

Change orders are a source of many disputes in today's construction industry. The issue at hand is whether or not the execution of change orders work has a negative impact on overall labor efficiency on a construction project. Previous literature demonstrates evidence that change orders affect labor efficiency. Attempts have been made to quantify these impacts by many researchers, with limited success. Using the electrical construction industry, a research study has been conducted to quantify the impacts of change orders on labor efficiency. In this paper, results of hypothesis testing and regression analysis are presented. A linear regression model that estimates the loss of efficiency, based on a number of independent variables, is also presented. The independent variables used in this model are (1) qualitative and quantitative criteria used to determine whether projects are impacted by changes or not; (2) the estimate of change order hours for the project as a percentage of the original estimate of work hours; (3) the estimate of change order hours for the project; and (4) the total number of years that the project manager had worked in the construction industry. Additional projects were used to validate the model, with an average error rate of 5%. The results of this research study are useful for owners, construction managers, general contractors, and electrical specialty contractors, because they provide a means to estimate the impact of a change order under certain project conditions. This research also identifies factors, which, when understood and effectively managed, may be used to mitigate the impact of a change order on project costs and efficiency.     

Change Orders Impact on Labor Productivity

This paper describes a study conducted to investigate the impact of change orders on construction productivity and introduces a new neural network model for quantifying this impact. The study is based on a comprehensive literature review and a field investigation of projects constructed in Canada and the USA. The field investigation was carried out over a 6-month period and encompassed 33 actual cases of work packages and contracts. Factors contributing to the adverse effects of change orders on labor productivity are identified and a model presented earlier is expanded to account primarily for the timing of change orders, among other factors. The developed model, as well as four models developed by others, have been incorporated in a prototype software system to estimate the loss of labor productivity due to change orders. A numerical example is presented to demonstrate the use of the developed model, and illustrate its capabilities.     

Impact of Change Orders on Labor Efficiency for Mechanical Construction

Change orders have become an everyday occurrence in construction. It is widely accepted by both owners and contractors that change orders have an effect on the labor efficiency, but these effects are difficult to quantify and frequently lead to disputes. Data from 61 mechanical construction projects were collected to develop a statistical model that estimates the actual amount of labor efficiency lost due to the change orders. The input variables needed in the model are as follows: (1) The original estimated labor hours; (2) impact classification; (3) total estimated change hours; (4) number of change orders; and (5) the timing of changes. The results of this study show that impacted projects have a larger decrease in labor efficiency than unimpacted projects. Additionally, the later a change order occurs in the life of a project the more impact it will have on the labor efficiency. The results appear to be consistent with the intuition of experienced professionals. Although each project has unique characteristics, the resultant model provides owners and contractors with a baseline measure of lost labor efficiency.     

Cumulative Effect of Project Changes for Electrical and Mechanical Construction

Change is inevitable on construction projects, primarily because of the uniqueness of each project and the limited resources of time and money that can be spent on planning, executing, and delivering the project. Change clauses, which authorize the owner to alter work performed by the contractor, are included in most construction contracts and provide a mechanism for equitable adjustment to the contract price and duration. Even so, owners and contractors do not always agree on the adjusted contract price or the time it will take to incorporate the change. What is needed is a method to quantify the impact that the adjustments required by the change will have on the changed and unchanged work. Owners and our legal system recognize that contractors have a right to an adjustment in contract price for owner changes, including the cost associated with materials, labor, lost profit, and increased overhead due to changes. However, the actions of a contractor can impact a project just as easily as those of an owner. A more complex issue is that of determining the cumulative impact that single or multiple change orders may have over the life of a project. This paper presents a method to quantify the cumulative impact on labor productivity for mechanical and electrical construction resulting from changes in the project. Statistical hypothesis testing and correlation analysis were made to identify factors that affect productivity loss resulting from change orders. A multiple regression model was developed to estimate the cumulative impact of change orders. The model includes six significant factors, namely: Percent change, change order processing time, overmanning, percentage of time the project manager spent on the project, percentage of the changes initiated by the owner, and whether the contractor tracks productivity or not. Sensitivity analysis was performed on the model to study the impact of one factor on the productivity loss (%delta). The model can be used proactively to determine the impacts that management decisions will have on the overall project productivity. They may also be used at the conclusion of the project as a dispute resolution tool. It should be noted that every project is unique, so these tools need to be applied with caution.     

Reconciliation of Owner and Contractor Views in Heavy Construction Projects

Delay and loss of productivity are the two main types of damage experienced by the contractor when the owner issues a change order. Courts have recognized critical path method schedule analysis as the preferred method of identifying and quantifying critical delays. As for the inefficiency damages, there is no direct way of measuring inefficiency due to its qualitative nature and the difficulty of linking the cause of the productivity loss to the damage. Most of the scholarly work published in this area was based on data supplied by the contractors; and that explains why there are discrepancies between what the contractor asks for and what the owner believes the contractor is entitled to. This study addresses the need for a statistical model to quantify the productivity loss from verifiable site data such as owner’s daily reports, change orders, drawings, and specifications, rather than rely solely on contractor surveys. A model is developed and validated to quantify the productivity loss in pipe work in roadway projects due to the change orders. The productivity loss study analyzed two sets of data that include: (1) variables that predict which of the two parties, the owner and the contractor, contributed to the productivity loss; and (2) variables that predict, from the legal viewpoint, productivity losses which only the owner is responsible for. The study showed the difference between what the contractor asked for and what he/she is actually entitled to. This model can be used by both the owner and the contractor to quantify the productivity loss due to change orders, and to offer an objective approach to reconcile their differences. This study concludes with an example to demonstrate the use of the model.     

A fuzzy adaptive learning control network with on-line structure and parameter learning

This paper addresses a general connectionist model, called Fuzzy Adaptive Learning Control Network (FALCON), for the realization of a fuzzy logic control system. An on-line supervised structure/parameter learning algorithm is proposed for constructing the FALCON dynamically. It combines the backpropagation learning scheme for parameter learning and the fuzzy ART algorithm for structure learning. The supervised learning algorithm has some important features. First of all, it partitions the input state space and output control space using irregular fuzzy hyperboxes according to the distribution of training data. In many existing fuzzy or neural fuzzy control systems, the input and output spaces are always partitioned into "grids". As the number of input/output variables increase, the number of partitioned grids will grow combinatorially. To avoid the problem of combinatorial growing of partitioned grids in some complex systems, the proposed learning algorithm partitions the input/output spaces in a flexible way based on the distribution of training data. Second, the proposed learning algorithm can create and train the FALCON in a highly autonomous way. In its initial form, there is no membership function, fuzzy partition, and fuzzy logic rule. They are created and begin to grow as the first training pattern arrives. The users thus need not give it any a priori knowledge or even any initial information on these. In some real-time applications, exact training data may be expensive or even impossible to obtain. To solve this problem, a Reinforcement Fuzzy Adaptive Learning Control Network (RFALCON) is further proposed. The proposed RFALCON is constructed by integrating two FALCONs, one FALCON as a critic network, and the other as an action network. By combining temporal difference techniques, stochastic exploration, and a proposed on-line supervised structure/parameter learning algorithm, a reinforcement structure/parameter learning algorithm is proposed, which can construct a RFALCON dynamically through a reward/penalty signal. The ball and beam balancing system is presented to illustrate the performance and applicability of the proposed models and learning algorithms.     

Predicting Industrial Construction Labor Productivity Using Fuzzy Expert Systems

The objective of this technical note is to illustrate the application of fuzzy expert systems to the modeling of a practical problem—that of predicting the labor productivity of two common industrial construction activities: rigging pipe and welding pipe. This note illustrates how to develop and test such a model, given the realistic constraints of subjective assessments, multiple contributing factors, and limitations on data sets. The factors that affect the productivity of each activity are identified, and fuzzy membership functions and expert rules are developed. The models are validated using data collected from an actual construction project. The resulting models are found to have high linguistic prediction accuracies. This note is of relevance to researchers by demonstrating how a fuzzy expert system can be developed and tested. It is of relevance to industry practitioners by illustrating how fuzzy logic and expert systems modeling can be exploited to help them solve real world problems.     

Impact of Change Orders on Small Labor-Intensive Projects

In today’s construction, small projects can be just as important if not more important than the larger projects. However, small projects are usually fast track projects, which often involve overlapping design and construction time. Subsequent modifications may be required for the sections that are already under construction. These disruptions to the ongoing project are labeled as change orders. The impact due to changes has been described as the adverse effect upon the unchanged work due to changes in the contract. For this study, 34 projects were selected to develop a statistical model that estimates the amount of labor efficiency lost due to change orders for small projects. The variables in the final model are percent design related changes, percent owner initiated changes, the ratio of actual peak labor to estimated peak labor, the ratio of actual project duration to estimated project duration, and project manager’s percent time on the project. The results of this paper are of value to owners, electrical and mechanical contractors, and construction managers. The model quantifies the impact of change orders by introducing the most important variables that bring the largest disruptions.     

Quantitative Definition of Projects Impacted by Change Orders

Change, defined as any event that results in a modification of the original scope, execution time, or cost of work, is inevitable on most construction projects due to the uniqueness of each project and the limited resources of time and money available for planning. Change may occur on a project for a number of reasons, such as design errors, design changes, additions to the scope, or unknown conditions. For each change, contractors are entitled to an equitable adjustment to the base contract price and schedule for all productivity impacts associated with the change. Changes may or may not have an impact on labor productivity. Existing literature uses subjective evaluation to determine whether the project is impacted. Projects impacted by change cause the contractor to achieve a lower productivity level than planned. The focus of this paper is to quantify whether an electrical or mechanical project is impacted by a change order. Through statistical hypothesis testing, groups of factors that correlate with whether a project is impacted by change orders were identified and used to develop a quantitative definition of impact. Logistic regression techniques were used to develop models that predict the probability of a project being impacted. The results of this research show that percent change, type of trade, estimated and actual peak manpower, processing time of change, overtime, overmanning, and percent change related to design issues are the main factors contributing to the project impact.     

Long-Term Impact of Equipment Technology on Labor Productivity in the U.S. Construction Industry at the Activity Level

In general, U.S. industries have witnessed dramatic changes in core processes over the past 25 years. Well understood technological and managerial advances have allowed the manufacturing sector, for example, to steadily increase its productivity and its profit margins. Similar changes are far less well understood in construction. This research examines 200 construction activities for the effect of technology, specifically equipment technology, on their labor productivity from 1976 to 1998. During that time period, changes in equipment technology are measured using a technology index consisting of five technology change factors. Through analysis of variance and regression analyses, it is found that activities experiencing significant changes in equipment technology have witnessed substantially greater long-term improvements in labor productivity than those that have not experienced a change in equipment technology. This research also reveals that changes in (1) energy, (2) control, and (3) functional range are significantly and positively correlated with improvements in labor productivity.     

Application of Fuzzy Logic to Simulation for Construction Operations

This paper describes the application of fuzzy logic to discrete event simulation in dealing with uncertainties of construction operations. The uncertainties in the quantity of resources required to activate an activity are modeled with fuzzy sets in linguistic terms. The fuzzy logic if-then rule is built to control the activation of activities. The duration of the activity that varies with the quantities of resources involved is determined through the fuzzy logic rule-based model. The fuzzy logic control of activities is incorporated with the activity scanning simulation strategy to implement the fuzzy simulation system for construction operations. In addition, the fuzzy activity element is adopted in the graphical modeling process. Examples are given that illustrate uses of the fuzzy simulation system and the impact of flexible demand of resources on productivity.     

Impact of Overmanning on Mechanical and Sheet Metal Labor Productivity

This paper details the impacts of overmanning on labor productivity for labor-intensive trades, namely mechanical and sheet metal contractors. Overmanning, as used in the following research, is defined as an increase of the peak number of workers of the same trade over the actual average manpower used throughout the project. The paper begins by reviewing literature on the effects of overmanning on labor productivity. Via a survey to various contractors, data were collected from 54 mechanical and sheet metal projects located across the United States. Various statistical analysis techniques are then performed to determine quantitative relationship between overmanning and labor productivity. These techniques include the stepwise method, T-test, P-value tests, analysis of variance, and multiple regression analysis. The results indicate a 0–41% loss of productivity depending on the level of overmanning and the peak project manpower. Cross-validation is performed to validate the final model. Finally, a case study is provided to demonstrate the application of the model.     

Neural Network Model for Estimating Construction Productivity

The paper describes an approach developed to estimate construction productivity for concrete formwork tasks. The system utilizes artificial neural networks, historical information, and input from experienced superintendents employed by a leading construction general contractor. It also summarizes a study undertaken to determine factors that affect labor productivity, the survey conducted to collect relevant data, and the design, training, and implementation of artificial neural networks at the participating company. A number of alternative neural network structures were investigated, the adopted one was a three-layered network with a fuzzy output structure. It was found that this structure provided the most suitable model since much of the input was subjective. A brief overview of the computer implementations and the overall experience with the system development is also provided. The method was compared to an existing statistical model developed by the same contractor and was found to improve the quality of the estimates attained. A case study conducted in the context of a workshop with senior estimators is also presented.     

Sensitivity Analysis of Neural Networks in Spool Fabrication Productivity Studies

The back-propagation neural network (BPNN) has been researched and applied as a convenient decision-support tool in a variety of application areas in civil engineering. However, learning algorithms such as the BPNN do not give information on the effect of each input parameter or influencing variable upon the predicted output variable. The model's sensitivity to changes in its parameters is generally probed by testing the response of a mature network on various input scenarios. In this paper, the relationships between an output variable and an input parameter are sorted out based on the BPNN algorithm. The input sensitivity of the BPNN is defined in exact mathematical terms in light of both normalized and raw data. The difference between a BPNN and regression analysis of statistics is discussed, and the sophistication and superiority of the BPNN over regression analysis is further demonstrated in a case study based on a small data set. In addition, statistical analysis of input sensitivity based on Monte Carlo simulation enables the modeler to understand the rationale of a BPNN's reasoning and have preknowledge about the effectiveness of model implementation in a probabilistic fashion. The sensitivity analysis of the BPNN is successfully applied to analyze the labor production rate of pipe spool fabrication in a real industrial setting. Important aspects of the application, including problem definition, factor identification, data collection, and model testing based on real data, are discussed and presented.     

Model for Quantifying the Impact of Change Orders on Project Cost for U.S. Roadwork Construction

Change orders are very common in almost every construction project nowadays, often resulting in increases of 5–10% in the contract price. Understanding the consequences of such trends, several studies have attempted to quantify the impact of change orders on the project cost. Most of the studies aimed at the quantification of the change orders were sponsored by contractors’ organizations, where statistical models used to quantify the impact of the change orders on the project cost were based on data supplied by the contractors; a situation that can lead to owner-contractor disagreements related to the quantification method used. In addition, most of the studies tackled commercial and electromechanical work, and very rare studies tackled the field of heavy construction; a field that suffers from change orders because of errors and omissions, scope of work changes, or changes because of unforeseen conditions. This study addresses the need for a statistical model to quantify the increase of the contract price due to change orders in heavy construction projects in Florida. The model is based on data collected from 16 Florida DOT projects with contract values that ranged between $10–$25 million, and that encountered an increase in the contract price from 0.01 to 15%. Eleven variables were analyzed to test their impact on the cost of the change orders. The study concluded that most significant variables that impact the value of the change order, which are (1) the timing of the change order and (2) when the reason for issuing the change order is unforeseen conditions. Two regression models are developed and validated as follows: (1) a model to quantify the percentage increase in the contract price due to the change orders that increase the contract price from 0.01 to 5% and (2) a model to quantify the percentage increase in the contract price due to the change orders that increase the contract price from 5 to 15%. Those models will provide the owner with a retrospective or forward pricing of the change orders, and hence, allow the owner to estimate and utilize contingency amounts.     

基于粗糙商集的模糊隶属度集值统计算法

模糊隶属度无统一算法,定义存在分歧.根据模糊概念"内涵明确,外延不明确"的特点,定义隶属度为不同外延对内涵的从属程度.在信息系统中,概念的外延用对象表示,内涵由属性表示,由此提出了求解隶属度的新算法:由原始统计数据组成初始信息系统,用粗糙集理论求得其商集并构建集值信息系统;该集值信息系统对应的条件概率空间中的条件概率即为隶属度.广义上信息系统可分为信息系统(无决策属性)和目标信息系统(有决策属性)两类.隶属度也可分为两类:第一类外延对象为内涵属性本身值,如年龄对青年人的隶属度(信息系统);第二类外延对象为不同于内涵属性的另一属性值,如边坡工程安全系数对稳定状态的隶属度(目标信息系统).计算以上两个实例,前者与已有结论作对比验证,后者与函数选择、经典统计方法及贝叶斯公理作对比验证,可知结果可靠,算法可行.      

ANN and Fuzzy Logic Models for Simulating Event-Based Rainfall-Runoff

This study presents the development of artificial neural network (ANN) and fuzzy logic (FL) models for predicting event-based rainfall runoff and tests these models against the kinematic wave approximation (KWA). A three-layer feed-forward ANN was developed using the sigmoid function and the backpropagation algorithm. The FL model was developed employing the triangular fuzzy membership functions for the input and output variables. The fuzzy rules were inferred from the measured data. The measured event based rainfall-runoff peak discharge data from laboratory flume and experimental plots were satisfactorily predicted by the ANN, FL, and KWA models. Similarly, all the three models satisfactorily simulated event-based rainfall-runoff hydrographs from experimental plots with comparable error measures. ANN and FL models also satisfactorily simulated a measured hydrograph from a small watershed 8.44?km2 in area. The results provide insights into the adequacy of ANN and FL methods as well as their competitiveness against the KWA for simulating event-based rainfall-runoff processes.     

Impact of Extended Overtime on Construction Labor Productivity

This paper presents an analysis of the impacts of extended duration overtime on construction labor productivity. The results show a decrease in productivity as the number of hours worked per week increase and/or as project duration increases. The research focuses on labor intensive trades such as the electrical and mechanical trades. Overtime in this research is defined as the hours worked beyond the typical 40 h scheduled per week. The paper begins by presenting the effects of overtime and the need for an updated overtime productivity model. Data for the quantitative analysis was collected from 88 projects located across the United States by means of a questionnaire. Various statistical analysis techniques were performed to develop quantitative relationship curves, including multiple regression, P-value tests, and analysis of variance.     

Genetic-Fuzzy Approach for Modeling Complex Systems with an Example Application in Masonry Bond Strength Prediction

A genetic-fuzzy learning from examples (GFLFE) approach is presented for determining fuzzy rule bases generated from input/output data sets. The method is less computationally intensive than existing fuzzy rule base learning algorithms as the optimization variables are limited to the membership function widths of a single rule, which is equal to the number of input variables to the fuzzy rule base. This is accomplished by primary width optimization of a fuzzy learning from examples algorithm. The approach is demonstrated by a case study in masonry bond strength prediction. This example is appropriate as theoretical models to predict masonry bond strength are not available. The GFLFE method is compared to a similar learning method using constrained nonlinear optimization. The writers’ results indicate that the use of a genetic optimization strategy as opposed to constrained nonlinear optimization provides significant improvement in the fuzzy rule base as indicated by a reduced fitness (objective) function and reduced root-mean-squared error of an evaluation data set.