# Forecasting Construction Cost Index through Artificial Intelligence

^{1}

^{2}

^{3}

^{4}

^{5}

^{*}

## Abstract

**:**

## 1. Introduction and Background

## 2. Research Methodology

#### 2.1. Data Sources

#### 2.2. The Formula for Calculating CCI

^{3}/PKR in the CCI (GW):

- BP
_{b}: Production value of bricks at the base year in PKR. - SP
_{b}: Production value of steel at the base year in PKR. - CP
_{b}: Production value of cement at the base year in PKR. - GP
_{b}: Production value of gravel and sand at the base year in PKR. - BU
_{b}: Unit rate of bricks at the base year in PKR/1000 bricks. - SU
_{b}: Unit rate of steel at the base year in PKR/ton. - CU
_{b}: Unit rate of cement at the base year in PKR/ton. - GU
_{b}: Unit rate of gravel and sand at the base year in PKR/m^{3}.

#### 2.3. CCI Prediction Techniques

#### 2.3.1. ANN

^{®}(Microsoft Corporation, Redmond, Washington, DC, USA). The calculation process involves two steps: studying and testing the algorithms. To achieve this, the data from 2000–2020 were divided into two parts, with 80% of the data being utilized for training and the remaining 20% being used to test the model.

- Procedure ANN

- 2
- Input ← database with all possible variable combinations

- 3
- for input = 1 to end of input, do
- 4
- for neurons = 1 to 10 do
- 5
- for repeat = 1 to 10 do
- 6
- Train ANN
- 7
- ANN storage ← save the highest test R
^{2} - 8
- end for
- 9
- end for
- 10
- ANN storage ← save the best prediction ANN depending on the input
- 11
- end for
- 12
- return ANN-Storage → Library with best predicting ANN for every variable combination
- 13
- end procedure

#### 2.3.2. Linear Regression

^{2}value.

#### 2.3.3. Time Series

- p = model’s order
- ${\varnothing}_{i}\dots \dots ..\text{}{\varnothing}_{p}$= model’s coefficients
- c = constant
- ${\epsilon}_{t}$ = white noise

^{3}) was used. This was selected as it provided the highest spatial autocorrelation. Following AR

^{3}, the highest accuracy was achieved at an order of 6.

## 3. Results, Analysis, and Discussions

- BW = 0.1192 in 1000 bricks/PKR
- SW = 0.0095 in tons/PKR
- CW = 0.0427 in tons/PKR
- GW = 0.126 in m
^{3}/PKR

- BU
_{i}: Unit rate of brick for the year i. - SU
_{i}: Unit rate of steel for the year i. - CU
_{i}: Unit rate of cement for the year i. - GU
_{i}: Unit rate of gravel and sand for the year i.

#### 3.1. Calculating and Forecasting CCI

#### 3.1.1. Calculating and Forecasting CCI Using ANN

#### 3.1.2. Calculating and Forecasting CCI Using the Linear Regression Method

^{2}value of 0.9269, indicating dependable outcomes. Figure 3 shows the predicted CCI values for 2021–2025, which display a nearly linear trend. However, this trend is unexpected given the continuous fluctuations in the historical CCI record for the years 2000–2020, influenced by the ever-changing political and security situation in Pakistan. An entirely linear forecast of future economic conditions seems unrealistic, and thus, the linear regression technique only provides an approximate trend for the predicted future CCI, which may not be as reliable as the predictions made by the ANN model.

#### 3.1.3. Calculating and Forecasting CCI Using Time Series

#### 3.2. Assessment of Forecasting Precision

## 4. Conclusions

## Author Contributions

## Funding

## Institutional Review Board Statement

## Informed Consent Statement

## Data Availability Statement

## Conflicts of Interest

## Abbreviations

ANN | Artificial Neural Network |

AR^{3} | Autoregressive model of order 3 |

ARIMA | Autoregressive integrated moving average (ARIMA) |

BW | Weight of Bricks |

CCI | construction cost index |

CW | Weight of Cement |

ENR | Engineering News Record |

GW | Weight of Gravel |

IMF | International Monetary Fund |

K-NN | K-Nearest Neighbors algorithm |

LSSVM | Least Squares SVM |

ME | Mean Error |

MAE | Mean Absolute Error |

PBS | Pakistan Bureau of Statistics |

PERT | Program Evaluation and Review Technique (PERT) |

PKR | Pakistani Rupees (PKR) |

ReLU | Rectified Linear Unit (ReLU) |

SSRIM | Self-adaptive structural radial basis neural network intelligence |

SVM | Support Vector Machine |

SW | Weight of Steel |

U-Stat | Theil’s U statistic |

VECM | Vector Error Correction Model (VECM) |

## References

- Marandi Alamdari, A.; Jabarzadeh, Y.; Samson, D.; Sanoubar, N. Supply chain risk factors in green construction of residential mega projects–interactions and categorization. Eng. Constr. Archit. Manag.
**2023**, 30, 568–597. [Google Scholar] [CrossRef] - Chen, P.-H.; Nguyen, T.C. A BIM-WMS integrated decision support tool for supply chain management in construction. Autom. Constr.
**2019**, 98, 289–301. [Google Scholar] [CrossRef] - Cao, Y.; Ashuri, B. Predicting the volatility of highway construction cost index using long short-term memory. J. Manag. Eng.
**2020**, 36, 04020020. [Google Scholar] [CrossRef] - Elfahham, Y. Estimation and prediction of construction cost index using neural networks, time series, and regression. Alex. Eng. J.
**2019**, 58, 499–506. [Google Scholar] [CrossRef] - Cao, M.-T.; Cheng, M.-Y.; Wu, Y.-W. Hybrid computational model for forecasting Taiwan construction cost index. J. Constr. Eng. Manag.
**2015**, 141, 04014089. [Google Scholar] [CrossRef] - Saar, C.C.; Chuing, L.S.; Yusof, A.M.; Zakaria, R.; Chuan, T.M. Construction cost index: A case study in malaysia. In Proceedings of the IOP Conference Series: Materials Science and Engineering; IOP Publishing: Bristol, UK, 2019; p. 012059. [Google Scholar]
- Wong, J.M.; Chan, A.P.; Chiang, Y.H. Time series forecasts of the construction labour market in Hong Kong: The Box-Jenkins approach. Constr. Manag. Econ.
**2005**, 23, 979–991. [Google Scholar] [CrossRef] - Wong, E.; Swei, O. New Construction Cost Indices to Improve Highway Management. J. Manag. Eng.
**2021**, 37, 04021030. [Google Scholar] [CrossRef] - Johnston, R.; Jones, K.; Manley, D. Confounding and collinearity in regression analysis: A cautionary tale and an alternative procedure, illustrated by studies of British voting behaviour. Qual. Quant.
**2018**, 52, 1957–1976. [Google Scholar] [CrossRef] - Holm, L.; Schaufelberger, J.E. Construction Cost Estimating; Routledge: London, UK, 2021. [Google Scholar]
- Lundberg, S.M.; Lee, S.-I. A unified approach to interpreting model predictions. In Proceedings of the 31st International Conference on Neural Information Processing Systems, Long Beach, CA, USA, 4–9 December 2017; pp. 4768–4777. [Google Scholar]
- Büyükşahin, Ü.Ç.; Ertekin, Ş. Improving forecasting accuracy of time series data using a new ARIMA-ANN hybrid method and empirical mode decomposition. Neurocomputing
**2019**, 361, 151–163. [Google Scholar] [CrossRef] - Lee, C.; Won, J.; Lee, E.-B. Method for predicting raw material prices for product production over long periods. J. Constr. Eng. Manag.
**2019**, 145, 05018017. [Google Scholar] [CrossRef] - Amigó, J.M.; Hirata, Y.; Aihara, K. On the limits of probabilistic forecasting in nonlinear times series analysis. Chaos: Interdiscip. J. Nonlinear Sci.
**2016**, 26, 123114. [Google Scholar] [CrossRef] [PubMed] - Zhang, N.; Lin, A.; Shang, P. Multidimensional k-nearest neighbor model based on EEMD for financial time series forecasting. Phys. A Stat. Mech. Its Appl.
**2017**, 477, 161–173. [Google Scholar] [CrossRef] - Ashuri, B.; Lu, J. Time series analysis of ENR construction cost index. J. Constr. Eng. Manag.
**2010**, 136, 1227–1237. [Google Scholar] [CrossRef] - Chakraborty, D.; Elhegazy, H.; Elzarka, H.; Gutierrez, L. A novel construction cost prediction model using hybrid natural and light gradient boosting. Adv. Eng. Inform.
**2020**, 46, 101201. [Google Scholar] [CrossRef] - Verm, A.; Verma, I. Use of artificial neural network in design of fly ash blended cement concrete mixes. Int. J. Recent Technol. Eng.
**2019**, 8, 4222–4233. [Google Scholar] [CrossRef] - Shiha, A.; Dorra, E.M.; Nassar, K. Neural networks model for prediction of construction material prices in Egypt using macroeconomic indicators. J. Constr. Eng. Manag.
**2020**, 146, 04020010. [Google Scholar] [CrossRef] - Williams, T.P. Predicting changes in construction cost indexes using neural networks. J. Constr. Eng. Manag.
**1994**, 120, 306–320. [Google Scholar] [CrossRef] - Kim, G.-H.; An, S.-H.; Kang, K.-I. Comparison of construction cost estimating models based on regression analysis, neural networks, and case-based reasoning. Build. Environ.
**2004**, 39, 1235–1242. [Google Scholar] [CrossRef] - Kim, B.-C.; Kwak, Y.H. Improving the accuracy and operational predictability of project cost forecasts: An adaptive combination approach. Prod. Plan. Control
**2018**, 29, 743–760. [Google Scholar] [CrossRef] - Fan, M.; Sharma, A. Design and implementation of construction cost prediction model based on SVM and LSSVM in industries 4.0. Int. J. Intell. Comput. Cybern.
**2021**, 14, 145–157. [Google Scholar] [CrossRef] - Wang, J.; Ashuri, B. Predicting ENR construction cost index using machine-learning algorithms. Int. J. Constr. Educ. Res.
**2017**, 13, 47–63. [Google Scholar] [CrossRef] - Moon, T. Forecasting construction cost index using interrupted time-series. KSCE J. Civ. Eng.
**2018**, 22, 1626–1633. [Google Scholar] [CrossRef] - Maqsoom, A.; Zulqarnain, M.; Irfan, M.; Ullah, F.; Alqahtani, F.K.; Khan, K.I.A. Drivers of, and Barriers to, the Adoption of Mixed Reality in the Construction Industry of Developing Countries. Buildings
**2023**, 13, 872. [Google Scholar] [CrossRef] - Ullah, F.; Thaheem, M.J.; Sepasgozar, S.M.; Forcada, N. System dynamics model to determine concession period of PPP infrastructure projects: Overarching effects of critical success factors. J. Leg. Aff. Disput. Resolut. Eng. Constr.
**2018**, 10, 04518022. [Google Scholar] [CrossRef] - Kamal, A.; Abas, M.; Khan, D.; Azfar, R.W. Risk factors influencing the building projects in Pakistan: From perspective of contractors, clients and consultants. Int. J. Constr. Manag.
**2022**, 22, 1141–1157. [Google Scholar] [CrossRef] - Ahmed, A.; Mehdi, M.R.; Baig, M.A.U.; Arsalan, M. The assessment of sustainability of freight transportation in Pakistan. Iran. J. Sci. Technol. Trans. Civ. Eng.
**2022**, 46, 2593–2608. [Google Scholar] [CrossRef] - Maqsoom, A.; Khan, M.U.; Khan, M.T.; Khan, S.; Ullah, F. Factors influencing the construction time and cost overrun in projects: Empirical evidence from Pakistani construction industry. In Proceedings of the 21st International Symposium on Advancement of Construction Management and Real Estate, The University of Hong Kong, Hong Kong, China, 14–16 December 2016; pp. 769–778. [Google Scholar]
- Iftikhar, R.; Müller, R.; Ahola, T. Crises and Coping Strategies in Megaprojects: The Case of the Islamabad–Rawalpindi Metro Bus Project in Pakistan. Proj. Manag. J.
**2021**, 52, 394–409. [Google Scholar] [CrossRef] - Ullah, F.; Al-Turjman, F. A conceptual framework for blockchain smart contract adoption to manage real estate deals in smart cities. Neural Comput. Appl.
**2023**, 35, 5033–5054. [Google Scholar] [CrossRef] - Kim, S.; Abediniangerabi, B.; Shahandashti, M. Pipeline construction cost forecasting using multivariate time series methods. J. Pipeline Syst. Eng. Pract.
**2021**, 12, 04021026. [Google Scholar] [CrossRef] - Kim, S.; Choi, C.-Y.; Shahandashti, M.; Ryu, K.R. Improving accuracy in predicting city-level construction cost indices by combining linear ARIMA and nonlinear ANNs. J. Manag. Eng.
**2022**, 38, 04021093. [Google Scholar] [CrossRef] - Shrestha, K. Cost Comparison of Highway Rest Area Operations: In-House Workforce versus Outsourcing Methods. J. Constr. Eng. Manag.
**2023**, 149, 04023062. [Google Scholar] [CrossRef] - Grogan, T. Dramatic drop in fuel prices starts to affect the markets. Eng. News Rec.
**2015**, 1, 1–16. [Google Scholar] - Khan, N.A.; Monti, G.; Nuti, C.; Vailati, M. Effects of Infills in the Seismic Performance of an RC Factory Building in Pakistan. Buildings
**2021**, 11, 276. [Google Scholar] [CrossRef] - Tufail, M.; Ahmad, N.; Mirza, S.M.; Mirza, N.M.; Khan, H.A. Natural radioactivity from the building materials used in Islamabad and Rawalpindi, Pakistan. Sci. Total Environ.
**1992**, 121, 283–291. [Google Scholar] [CrossRef] - Ramachandran, P.; Zoph, B.; Le, Q.V. Searching for activation functions. arXiv
**2017**, arXiv:1710.05941. [Google Scholar] - Prajit, R.; Zoph, B.; Quoc, V.L. Swish: A self-gated activation function. arXiv
**2017**, arXiv:1710.059417. [Google Scholar] - Ahmad, M.W.; Mourshed, M.; Rezgui, Y. Trees vs Neurons: Comparison between random forest and ANN for high-resolution prediction of building energy consumption. Energy Build.
**2017**, 147, 77–89. [Google Scholar] [CrossRef] - Wu, N.; Green, B.; Ben, X.; O’Banion, S. Deep transformer models for time series forecasting: The influenza prevalence case. arXiv
**2020**, arXiv:2001.08317. [Google Scholar] - Atif, S.; Umar, M.; Ullah, F. Investigating the flood damages in Lower Indus Basin since 2000: Spatiotemporal analyses of the major flood events. Nat. Hazards
**2021**, 108, 2357–2383. [Google Scholar] [CrossRef] - Adeli, H.; Wu, M. Regularization neural network for construction cost estimation. J. Constr. Eng. Manag.
**1998**, 124, 18–24. [Google Scholar] [CrossRef] - Bode, J. Neural networks for cost estimation: Simulations and pilot application. Int. J. Prod. Res.
**2000**, 38, 1231–1254. [Google Scholar] [CrossRef] - Aslam, A.; Rana, I.A.; Bhatti, S.S. The spatiotemporal dynamics of urbanisation and local climate: A case study of Islamabad, Pakistan. Environ. Impact Assess. Rev.
**2021**, 91, 106666. [Google Scholar] [CrossRef] - Jafarzadeh, R.; Ingham, J.M.; Wilkinson, S.; González, V.; Aghakouchak, A.A. Application of artificial neural network methodology for predicting seismic retrofit construction costs. J. Constr. Eng. Manag.
**2014**, 140, 04013044. [Google Scholar] [CrossRef] - Wilmot, C.G.; Mei, B. Neural network modeling of highway construction costs. J. Constr. Eng. Manag.
**2005**, 131, 765–771. [Google Scholar] [CrossRef] - Wang, Y.-R.; Yu, C.-Y.; Chan, H.-H. Predicting construction cost and schedule success using artificial neural networks ensemble and support vector machines classification models. Int. J. Proj. Manag.
**2012**, 30, 470–478. [Google Scholar] [CrossRef]

Year | CCI (%) | Year | CCI (%) |
---|---|---|---|

2000 | 719 | 2011 | 1090 |

2001 | 756 | 2012 | 1180 |

2002 | 803 | 2013 | 1170 |

2003 | 840 | 2014 | 1185 |

2004 | 880 | 2015 | 1260 |

2005 | 775 | 2016 | 1275 |

2006 | 890 | 2017 | 1299 |

2007 | 975 | 2018 | 1341 |

2008 | 950 | 2019 | 1336 |

2009 | 998 | 2020 | 1345 |

2010 | 1022 |

Year | ANN | Linear Regression | Time Series |
---|---|---|---|

2021 | 1405 | 1423 | 1376 |

2022 | 1462 | 1467 | 1444 |

2023 | 1509 | 1487 | 1517 |

2024 | 1540 | 1529 | 1545 |

2025 | 1533 | 1566 | 1506 |

Error | ANN | Regression | Time Series |
---|---|---|---|

ME | 0.04 | 0.3 | 0.22 |

RMSE | 34.91 | 36.19 | 36.82 |

MAE | 28.3 | 28.3 | 30.07 |

U-Stat | 0.62 | 0.64 | 0.65 |

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |

© 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Aslam, B.; Maqsoom, A.; Inam, H.; Basharat, M.u.; Ullah, F.
Forecasting Construction Cost Index through Artificial Intelligence. *Societies* **2023**, *13*, 219.
https://doi.org/10.3390/soc13100219

**AMA Style**

Aslam B, Maqsoom A, Inam H, Basharat Mu, Ullah F.
Forecasting Construction Cost Index through Artificial Intelligence. *Societies*. 2023; 13(10):219.
https://doi.org/10.3390/soc13100219

**Chicago/Turabian Style**

Aslam, Bilal, Ahsen Maqsoom, Hina Inam, Mubeen ul Basharat, and Fahim Ullah.
2023. "Forecasting Construction Cost Index through Artificial Intelligence" *Societies* 13, no. 10: 219.
https://doi.org/10.3390/soc13100219