Tiling Optimization

Here’s a common problem in fabrication and construction: how much material should we buy? Anyone who’s tried to do a DIY home project or just make something physical knows the answer is, “more than you think you need.” But in commercial construction, where there are formal schedules and budgets that are part of a contract, it can be helpful to quantify that answer. But since you can’t know the real answer here until you’ve finished the project, some modeling and estimation is needed.

That’s the premise of this study on a custom floor tiling project for a restaurant dining room. Both the shape of the floor plan and the tiles themselves are irregular. But the real question is, How do we account for partial tiles in our estimation and how should that inform our material takeoff?

The approach can be thought of as an algorithm with four steps and, indeed, was implemented as one. Those steps were: 1. Count the full tiles, 2. Establish a cutoff for the remaining partial tiles, based on area, to classify them for the model, 3. Add up the total material area of the partials, and 4. Find the difference between the total area of the estimated partial tiles and the equivalent number of full tiles. The goal of the process is to minimize the difference between these two values by manipulating the area cutoff point.

The result was to move the initial contractor estimate for a 25-30% contingency down significantly, by offering a data-driven estimate of material overage.