This research project proposed a two-step coordination approach that contractors should take to help both suppliers and contractors responsively cope with the demand fluctuations with the minimal cost over the dynamic construction processes. The first step is to automatically calculate the demand fluctuations from a 4D model that monitors the as-built site progress and helps the contractor update the look-ahead schedules. The second step is to make decisions on the material orders at the occurrence of the demand fluctuation determined in the first step. The decisions on the material orders consist of extra-orders and outsourced orders, which are determined by the heuristics, the CMA-ES approach. As the coordination approach enables a data-driven and schedule-driven supply chain coordination process to address uncertain demands, it would contribute to the current body of literature not only in supply chain information visibility but also in supply chain responsiveness through efficient decisions made.

To test the usefulness of the proposed approach, an example office-building project featuring a five-day in-situ construction of concrete walls was tested to derive the revised order plan in a 5-day look-ahead planning window. The 4D model facilitated the data collection on the site progress monitoring and the quick computation of the demand fluctuation quantities. In the example demonstration, the optimal revised order plan was obtained to reduce the total cost by 4.3% when the proposed coordination approach was implemented instead of the traditional approach. This reduction should be viewed as a substantial improvement for the existing practice of supplier-contractor coordination.

The sensitivity analysis was conducted to explore the effects of parameters on the CMA-ES optimization results, i.e., the total costs from the revised order plans. An interesting finding from the sensitivity analysis was that the contractor should encourage the supplier to make the best use of idle time for production. It will allow enough extra-order quantities to be supplied to fulfill the increasing demand with the minimal total cost. Additionally, the magnitude of the delayed quantities to be carried forward and the unit cost for producing the outsourced quantities do not have a significant impact on the total cost. It is, however, necessary to investigate more parameters for the sensitivity analysis.

In addition, the supplier-contractor coordination approach was tested only for one example project featuring the in-situ wall installations, whereas the construction processes consist of multiple disciplines, very complex precedential relationships, and various scheduling constraints. These all have a potential influence on the practical implementation of the proposed approach. The approach can be extended to cover all possible constraints that may affect the determination of the optimal solutions for demand fluctuations. For example, the tasks on the critical path have the highest priority to fulfill the demand, which should be modeled into the cost minimization process. Also, the success of heuristic algorithms is highly dependent on the specific problem formulation case by case. The CMA-ES approach selected for this research should be further validated for its fitness for more complex problem settings. Apart from that, issues regarding the data interoperability, cybersecurity, legal and contractual relationships should be investigated to improve the practicality of the proposed approach.

link to the full research paper