Advanced data analytics ensures flawless management of construction projects

Working with big data analytics (BDA), this construction company has been able to minimise losses from late deliveries in its project schedule. Data analytics facilitated predictive abilities that became integral to the company’s new centralized system that now enabled different levels of executives to monitor, benchmark and act on a variety of factors such as manpower allocation, machinery placement, resource scheduling, up-to-date project progress and expenditures.

Challenges:
In the construction sector, good project schedule and cost estimation, on top of ensuring timely project execution are very important. Contractors and developers get penalized with Liquidated and Ascertained Damages or LADs – these are fixed damages stated in the building contract, and usually set as an amount per week or part of a week which the contractor must pay the employer if completion is delayed beyond the contractual date for completion – that sometimes amount to millions of Ringgit for late deliveries. Therefore, the need arose for a centralized system that will allow different levels of executives to monitor, benchmark, and act on a variety of factors. These factors include manpower allocation, machinery placement, resource scheduling, up-to-date project progress and expenditures, which both directly and indirectly contribute to desired project schedules and costs.

quandatics-construction-infographic

Prior to implementing Quandatics’ advanced analytics solution, data was scattered around in spreadsheets owned by different departments, namely Site Management, Project Management, Finance and Human Resources. This made it difficult for upper management to obtain a complete overview of important metrics without spending time getting people together to consolidate on-hand data.

The limited interactive-ness of old-fashioned spreadsheet visualization also constrained the amount of understanding all the users could obtain from available data. A lot of times, problems are only detected after weeks of ongoing events and consequently, the implemented corrective measures were much less effective.

Solution:
The first step toward a solution was the consolidation of historical datasets into a centralized database system, with mobile and web interfaces for supervisors to append up-to-date data of on-going projects into the system.  This allows for extraction of descriptive insights, and benchmarking to be performed on a number of important metrics that constitutes the project schedules, so that senior management can make data-driven decisions.

A key component in the current implementation is a project management portal for schedule, progress, and cost monitoring to optimize resources allocation. An interactive infographic service with multi-level access is provided for personnel with different access rights to visualize or append the data and results according to their job functions.

quandatics-construction

Project management infographics for monitoring individual project progress, resource usage and costing

Business Benefits:
Previously, manual work was required to consolidate and aggregate data from multiple departments in order to report the equivalent amount of information. It’s now instant, with consistently updated progress visualization for all on-going projects available on multiple devices and multiple access levels.  Furthermore, standardized reporting and progress update interface for all site supervisors are now enabled, eliminating scattered records and logs and potential loss of information. With these capabilities, the management team is now able to respond to potential deficiencies or problems much faster – from one week previously to within a day now.

The Future:
By consolidating data from past and current piling projects, correlations between their attributes and the project outcome, profits and schedules can be studied. Moving forward, more data sources, such as ground survey data and weather data can also be incorporated to make the system more comprehensive.

Future implementation with machine learning features will generate best-action recommendations for on-going projects. These outputs could potentially lead to the minimization of LADs, reduction of material wastage, and optimization of machinery placement, which are among the issues with the highest priorities in piling projects.

, , , , , ,