Graph-based Automated construction Scheduling (GAS) without the use of BIM - A case study

Oct 22, 2026

The construction industry faces long-standing delay and cost overrun driven by schedulers’ tacit experience that is rarely captured, stored, or reused. A graph-based automated scheduling (GAS) method is proposed to extract schedule features, classify sequences, assemble them into schedules, and optimize time-and cost-efficiency. Validation on two case studies shows the generated construction schedule is, on average, 6.70% closer to the actual schedule than the planned schedule. The approach requires limited BIM availability and helps initiate new schedules earlier.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages77

Reading Time3 h 14 min

AccessFree to download

Manuscript metadata

Abstract and classifications

Journal submission questions

Fit with aims and scope

Case study suitability

Novelty and prior submissions

Journal of Construction Engineering and Management Graph-based Automated construction Scheduling (GAS) without the use of BIM --Manuscript Draft-- | Manuscript Number: | COENG-12687R1 | | --- | --- | | Full Title: | Graph-based Automated construction Scheduling (GAS) without the use of BIM | | Manuscript Region of Origin: | AUSTRALIA | | Article Type: | Case Study | | Manuscript Classifications: | Artificial intelligence and machine learning; Pre-Project Planning; Scheduling | | Funding Information: | Innovate UK (104795) Not applicable | | Abstract: | The construction industry has been suffering from delay and cost overrun for decades. Experienced schedulers programs and allocate contingencies (both cost and time) based on professional experiences and gained knowledge. Such tacit knowledge has not been captured, stored, and shared with inexperienced schedulers. This paper proposes a Graph-based Automated Scheduling (GAS) method to capture, store andreuse the tacit knowledge in the construction schedules. The proposed GAS method takes construction schedules as input, extracts schedule features, classifies schedules into different types of sequences, selects and assembles sequences into schedules, and eventually optimises time-and cost-efficiency of assembled schedules. The GAS method is validated on two case studies. The results indicate that the automatically generated construction schedule is, on average, 6.70% closer to the actual schedule than the planned schedule . Different from existing automatic scheduling methods, GAS relies little on the availability and data richness of BIM models. Hence, GAS helps schedulers initiate new schedules more efficiently at the early construction stages | | Corresponding Author: | Ying Hong
University of Cambridge
Cambridgeshire, UNITED KINGDOM | | Corresponding Author E-Mail: | [hongying1525@gmail.com](hongying1525@gmail.com) | | Order of Authors: | Ying Hong | | | Haiyan Xie | | | Eva Agapaki | | | Ioannis Brilakis | | Suggested Reviewers: | | | Opposed Reviewers: | | | Additional Information: | | | Question | Response | | The journal requires that all submissions fall within its aims and scope, explained here. Please explain how your submission fits the journal's aims and scope. | This manuscript focuses on improving the accuracy of construction schedules. Therefore, it falls in the cost and schedule, as well as project planning topics. | | Papers published in ASCE Journals must make a contribution to the core body of knowledge and to the advancement of the field. Authors must consider how their new knowledge and/or innovations add value to the state of the art and/or state of the practice. Please outline the specific contributions of this research in the comments box. | The proposed GAS method is validated on a bridge project to automatically generate construction schedules without Building Information Models (BIM), and then compare with the original schedule regarding the time-and cost-efficiency. Different from existing automatic scheduling methods, GAS relies little on the availability and data richness of BIM models. Hence, GAS helps schedulers initiate new schedules more efficiently at the early construction stages. | Powered by Editorial Manager® and ProduXion Manager® from Aries Systems Corporation |

To read more about how the JCEM considers each paper’s contributions please see the following href="[http://dx.doi.org/10.1061/](http://dx.doi.org/10.1061/)(ASCE)CO.1943-7862.0000708"
target="_blank">Editor's Note . | | | --- | --- | | Case studies discuss an empirical theory, new practices/methods/tools, innovations or lessons learned, related to engineering, design, construction, and/or project management, discovered or defined using a qualitative but structured research methodology. Case studies are typically based on actual experience(s) and/or observations on a single or multiple reallife applications of a particular topic in the construct

FAQs

What problem does the proposed GAS method address in construction scheduling?

It targets delays and cost overruns caused by the lack of captured and reusable tacit knowledge from experienced schedulers.

How does GAS generate a construction schedule?

GAS takes existing schedules as input, extracts schedule features, classifies them into sequence types, assembles selected sequences into schedules, and optimizes time-and cost-efficiency.

How is GAS different from existing automatic scheduling methods regarding BIM?

GAS relies little on the availability and data richness of BIM models, enabling schedule initiation earlier in construction stages.

Oct 22, 2026

Uploaded by Emma Wilson