Material Selection Chart Method

This Python Notebook code is a tool to select material from a database using the chart and material index method

🔰 How does it work?

• A material database is uploaded. In this case the database was created by my students from my Selection Material Course using online free database as Matweb and Azom
• Material database has the below information:
  • Material's name
  • Family (Ceramics, Metals, Polymers, Hybrids)
  • Price (USD/kg)
  • Density (kg/m^3)
  • Young's Modulus (GPa)
  • Yield Strength (MPa)
  • Ultimate Strength (MPa)
  • Elongation (%)
  • Hardness (HV)
  • Fracture Toughness [MPa m^1/2]
  • Melting Point
  • Source (Matweb, Azom, etc)
  • Person (who include the material information into the database)

• Source and Person columns are deleted from the database
• A bar plot (Family) and histograms for the material's properties are showed

• User has to choose two properties to create the material index. This is an example choosing Young's Modulus and Density

• User must enter the exponents for each property. Based on this, the material index is created. In this example 0.5 and -1 are the exponents for Young's Modulus and Density, respectively. With these values, the material index is: $$\frac{E^{0.5}}{\rho}$$
• Selection Orientation is defined for the filtering process, based on the properties' exponents.
• A maximization would be performed if the exponent is positive
• A minimization would be performed if the exponent is negative
• Material Index is calculated for each materila into the database
• Used must set the number of material to filtering
• The list of materials sort by mateial index is showed

• A material chart is showed
• First property as Y-Axis and second property as X-axis
• Each point represents a material and its color depends of the family
• A black line represents the filter based on the number of material that user wants to classify
• The green area is the selection region
• The gray area is the discard region

• It is possible to show a chart with the cluster families

🔶 What is next?

• Increase the database with more materials and properties
• Create interactive charts
• Create an interface or app