Introduction

The study arose from the need to easily analyze data provided by a local supplier of structural steel members. The data is contained in an Excel spreadsheet as a price list, but in a not tidy arrangement difficult to analize. Using R software together with the interaction power of Shiny, it allows the user to analyze the price distribution of structural components to make design related decisions.

The particular interest lies in obtaining the total costs of the components with respect to their weight.

Data Analysis

Extraction / Transformation

The process starts by the accessing the supplier’s file. Once the data has been read, we proceed to:

• Tidy up the data previously arranged in blocks of materials categories. A tidy structure is generated by arranging the blocks as category columns.
• The data is filtered and only the categories of interest are left.
• The necessary information is extracted from each of the descriptions of the elements by means of regular expression programming. Data as dimensions, weight, bar length, etc. are extracted.

Processing

Once data is ready, a few calculations are performed before analisys:

• The cost in Argentinian Pesos is Dollarized to the official exchange rate of the price list date
• The cost per unit weight (u$s/kg) is calculated for each member.
• A Shiny App is created to perform the data exploration.
• Frequency Plots, BoxPlots and ScatterPlots are included in the graphs.

The following graphic shows an image taken from the Shiny app.

knitr::include_graphics("Freqplot.png")

References & links

Access the Shiny App:

Shiny App

Or access the scripts and data files through GitHub.

Git Hub / Sergio Miranda / Structural Steel Cost Analysis

If not using Shiny App run the Shiny scripts locally in R Studio to access the interactive graphics.