ANN Model to Predict Final Construction Contract Duration: History
View Latest Version
Please note this is an old version of this entry, which may differ significantly from the current revision.
Contributor:
Abdullah M. Alsugair
,
Khalid S. Al-Gahtani
,
Naif M. Alsanabani
,
Abdulmajeed A. Alabduljabbar
,
Abdulmohsen S. Almohsen

Forecasting the final construction contract duration at an early stage plays a vital role in the progress of a project. An inaccurate project duration prediction may lead to the project’s benefits being lost. It is essential to precisely predict the duration due to the presence of several different factors.

  • duration
  • contract
  • logarithmic
  • transformed
  • cost
  • sector
  • errors

1. Introduction

The construction industry significantly contributes to countries’ economic progress. Construction is an industry that contributes significantly to the overall Gross Domestic Product (GDP) and is expected to expand. Delays in construction projects have become a widespread issue due to the complexity of the construction industry. Despite having a positive impact on the economy and technological improvements in the sector [1][2], construction delays have a wide range of social and economic repercussions. These delays negatively impact sustainability’s social, environmental, and financial triple bottom lines [3]. Delays can lead to schedule and cost overruns, decreased contractor earnings, additional losses for the owner’s capital due to an extended construction phase, mistrust between the owner and contractor, legal battles involving many parties, and outright project abandonment. Gebrehiwet and Luo [4] noted that cost overruns, contract cancellation, arbitration, and litigation are some of a delay’s crucial effects. According to Khattri et al. [5], a delay can result in disagreements, cost overruns, time overruns, abandonment, negotiation, legal action, litigation, and complete desertion. Numerous studies have been carried out over the years to address this significant issue, especially to identify the underlying factors that increase the probability of a building delay and its adverse effects.
Researchers in multiple nations have revealed numerous harmful consequences of delays. Hecker [6] claimed that significant cost overruns ranging from 40% to 400% occurred on various infrastructure projects in the United States of America. In the United Kingdom of Great Britain, just 38% of projects were finished within 5% of the contract timeline program, and only 70% were completed within 5% of the tender cost [7]. Additionally, according to [8], only one-eighth of Australian building projects were finished by the deadlines outlined in the contracts, and the typical schedule overrun was greater than 40%. Moreover, Sodangi and Salman [9] stated that about 70% of projects in Nigeria encountered delays, demonstrating that delays are a significant issue in Nigerian construction. Only 30% of building projects in the Kingdom of Saudi Arabia (KSA) were finished on time, while the typical time overrun was up to 30% [10].
Most studies have focused on identifying and analyzing the factors affecting delays in projects [4][11][12][13]. They pointed out that the factors that cause poor project performance still need to be fully understood. The causes of time delays were different from region to region. There is a need for more studies to predict the final construction contract duration in the early stages to assist stakeholders in deciding whether to continue or halt the project.

2. Forecast Model Based on the Causes of the Time Delays

In this section, the researchers tried to find the significant factors that affected the time delay and integrated them as input data to forecast the final construction contract duration. In a general project, Al-Gahtani et al. [14] used ten previously discovered criteria to construct a simulation forecasting model for the delay duration in Saudi projects using system dynamics. In order to consider the ten factors that influence project delay, they carried out a systematic, integrated approach using the DEMATEL methodology and system dynamics (SD). This work solved the challenge of methodically creating a causal loop diagram inside the SD modeling process using the DEMATEL technique. Next, consistency and extreme conditions were tested on the generated SD model. Then, it was implemented and validated using three case studies in KSA by contrasting each case study’s real and fitted progress curves. In addition, Ajayi and Chinda [2] developed a model to examine the impact of the factors on the final construction project time. The model combined two mathematical decision-making techniques, DEMATEL and SD modeling. The simulation findings highlight the significance of avoiding design errors at the project’s beginning (or preconstruction stage) to reduce project delays.
For highway projects, Pewdum et al. [15] evolved models to project the final cost and duration of a highway construction project while it was still in the planning stages. Before designing the forecasting models, project data were gathered and examined to determine the variables influencing the project’s ultimate budget and duration. The research for these models was based on the ANN. Han et al. [16] examined the influence of the non-value-adding effort generated from design errors and changes in design on the time delay of the project using system dynamics.
In order to facilitate reliable project delay risk analysis and forecasting using objective data sources, Gondia et al. [17] refined and built machine learning (ML) algorithms (decision tree and naïve Bayesian algorithms). As a result, the relevant delay risk sources and components were first found. A multivariate data set of past project timeliness and delay-inducing risk sources was assembled. Exploratory data analysis was then used to reveal the system’s intricacy and interconnectivity. In order to anticipate the extent of project delays, the two appropriate algorithms were found and trained using the data set. These models used decision trees and naive Bayesian classification algorithms. Finally, cross-validation tests were performed on both models to assess their predictive abilities. The models were then contrasted using performance metrics pertinent to ML.

3. Forecast Model Based on Characteristics of a Contract or Project

Although the earned value management (EVM) approach is a successful project oversight and management strategy in terms of foretelling the cost performance index and other cost indicators, the technique may require more improvements to be more effective at estimating the project’s completion time [18][19]. Vanhoucke and Vandevoorde [19] assumed that project activities and precedence relations were known to predict the final contract duration (FCCD). Urgilés et al. [20] examined the adequate EVM and value schedule to forecast the final duration of hydroelectric power generation projects. Sackey et al. [21] also developed a new method based on the EVM to forecast the final construction contract duration (FCCD). They used the actual time spent on each activity. One of the challenges faced by the users of the management method in predicting the actual duration of the contract is that the method requires historical data for the project. In other words, EVM also needs accurate information from a project, such as its cost, earned value, and planned value, at any given time, and it may not be possible to predict it at an early stage of the contract.
On the other hand, the case-based reasoning method is mainly used to forecast the construction project cost. However, Jin et al. [22] established a CBR model that can correctly predict the FCCD at the planning stage.

4. Regression and ANN Models

Several studies utilized regression and ANN models to estimate the FCCD. For example, Skitmore and Ng [23] developed a regression model based on cross-validation. The model parameters were project type, sector, contractor selection, and the 93 Australian building project model. Thomas and Thomas [24] developed a regression model to predict the building project duration based on 51 historical data. The model parameters were the area of the building, estimated duration, and estimated cost. The model did not consider the project sectors, and the model cannot be used for different types of projects, such as electrical or mechanical projects. The artificial neural network method proved more advanced and performed better than the regression model [25]. The ANN model developed by [26] was to forecast the duration of building projects. The input data included the number of floors, foundation type, activities, contractor class and client class, and floor area. The mean absolute error was 25.9%. Moreover, Gab-Allah et al. [27] established an ANN model for predicting the building project. The parameters were the type of clients, construction quality, project location, the total height of the project, client coordination with contractor staff, contract type (unit price contract/lump sum), contactor selection method, and quality of project documentation. The maximum error of the model was 20%.
The previous models required specific information, which varies from one project to another, such as the model developed by Al-Gahtani et al. [14], Ajayi and Chinda [2], Pewdum et al. [15], or contract data that should be available through the construction stage, such as Sackey et al. [23]. Although the CBR method has proven its effectiveness in predicting the duration of the contract, it requires the availability of a previously completed project similar to the one required in terms of characteristics and operational conditions, which may be difficult to provide. In terms of the regression and ANN models, the above model was utilized for building project duration and cannot be generalized to other projects. Therefore, there is a need to develop a predictive model using the ANN model, which is used for different projects and is based on common and available data.

This entry is adapted from the peer-reviewed paper 10.3390/app13148078

References

  1. Bertelsen, S.; Sacks, R. Towards a new understanding of the construction industry and the nature of its production. In Proceedings of the 15th Conference of the International Group for Lean Construction, East Lansing, MI, USA, 18–20 July 2007; Michigan State University: East Lansing, MI, USA, 2007; pp. 46–56.
  2. Ajayi, B.O.; Thanwadee, C. Impact of construction delay controlling parameters on project schedule: DEMATEL-System dynamics modelling approach. Front. Built Environ. 2022, 8, 799314.
  3. Tafazzoli, M.; Shrestha, P.P. Investigating Causes of Delay in US Construction Projects. In Proceedings of the 53rd ASC Annual International Conference, Seattle, WA, USA, 5–8 April 2017.
  4. Gebrehiwet, T.; Luo, H. Analysis of Delay Impact on Construction Project Based on RII and Correlation Coefficient: Empirical Study. Procedia Eng. 2017, 196, 366–374.
  5. Khattri, T.; Agarwal, S.; Gupta, V.; Pandey, M. Causes and Effects of Delay in Construction Project. Int. Res. J. Eng. Technol. 2016. Available online: www.irjet.net (accessed on 4 March 2023).
  6. Hecker, J.Z. Transportation Infrastructure: Cost and Oversight Issues on Major Highway and Bridge Projects; US General Accounting Office: Washington, DC, USA, 2002.
  7. Latham, S.M. Constructing the Team; HMSO: London, UK, 1994.
  8. Bromilow, F.J. Measurement and scheduling of construction time and cost performance in the building industry. Chart. Build. 1974, 10, 57–65.
  9. Sodangi, M.; Salman, A. AHP-DEMATEL modelling of consultant related delay factors affecting sustainable housing construction in Saudi Arabia. Int. J. Constr. Manag. 2022.
  10. Assaf, S.A.; Al-Hejji, S. Causes of delay in large construction projects. Int. J. Proj. Manag. 2006, 24, 349–357.
  11. Al Saeedi, A.S.; Karim, A.M. Major Factors of Delay in Developing Countries Construction Projects: Critical Review. Int. J. Acad. Res. Bus. Soc. Sci. 2022, 12, 797–809.
  12. Sambasivan, M.; Soon, Y.W. Causes and effects of delays in Malaysian construction industry. Int. J. Proj. Manag. 2007, 25, 517–526.
  13. Love, P.; Edwards, D.J. Forensic project management: The underlying causes of rework in construction projects. Civ. Eng. Environ. Syst. 2004, 21, 207–228.
  14. Al-Gahtani, K.; Alsugair, A.; Alsanabani, N.; Alabduljabbar, A.; Almutairi, B. Forecasting delay-time model for Saudi construction projects using DEMATEL–SD technique. Int. J. Constr. Manag. 2022, 1–15.
  15. Pewdum, W.; Rujirayanyong, T.; Sooksatra, V. Forecasting final budget and duration of highway construction projects. Eng. Constr. Archit. Manag. 2009, 16, 544–557.
  16. Han, S.; Lee, S.; Peña-Mora, F. Identification and Quantification of Non-Value-Adding Effort from Errors and Changes in Design and Construction Projects. J. Constr. Eng. Manag. 2012, 138, 98–109.
  17. Gondia, A.; Siam, A.; El-Dakhakhni, W.; Nassar, A.H. Machine learning algorithms for construction projects delay risk prediction. J. Constr. Eng. Manag. 2020, 146, 4019085.
  18. Anbari, F.T. Earned value project management method and extensions. Proj. Manag. J. 2003, 34, 12–23.
  19. Vanhoucke, M.; Vandevoorde, S. A simulation and evaluation of earned value metrics to forecast the project duration. J. Oper. Res. Soc. 2007, 58, 1361–1374.
  20. Urgilés, P.; Claver, J.; Sebastián, M.A. Analysis of the earned value management and earned schedule techniques in complex hydroelectric power production projects: Cost and time forecast. Complexity 2019, 2019, 3190830.
  21. Sackey, S.; Lee, D.E.; Kim, B.S. Duration Estimate at Completion: Improving Earned Value Management Forecasting Accuracy. KSCE J. Civ. Eng. 2020, 24, 693–702.
  22. Jin, R.; Han, S.; Hyun, C.; Cha, Y. Application of Case-Based Reasoning for Estimating Preliminary Duration of Building Projects. J. Constr. Eng. Manag. 2016, 142, 04015082.
  23. Skitmore, R.M.; Ng, S.T. Forecast models for actual construction time and cost. Build. Environ. 2003, 38, 1075–1083.
  24. Thomas, N.; Thomas, A.V. Regression modelling for prediction of construction cost and duration. Appl. Mech. Mater. 2017, 857, 195–199.
  25. Badawy, M. A hybrid approach for a cost estimate of residential buildings in Egypt at the early stage. Asian J. Civ. Eng. 2020, 21, 763–774.
  26. Ujong, J.A.; Mbadike, E.M.; Alaneme, G.U. Prediction of cost and duration of building construction using artificial neural network. Asian J. Civ. Eng. 2022, 23, 1117–1139.
  27. Gab-Allah, A.A.; Ibrahim, A.H.; Hagras, O.A. Predicting the construction duration of building projects using artificial neural networks. Int. J. Appl. Manag. Sci. 2015, 7, 123–141.
More
© Text is available under the terms and conditions of the Creative Commons Attribution (CC BY) license; additional terms may apply. By using this site, you agree to the Terms and Conditions and Privacy Policy.
This entry is offline, you can click here to edit this entry!
ScholarVision Creations
Feedback