B-5 Relational Database Theory - Arthur McCloskey

     A relational database is a collection of data items that are organized in a manner that relates the parameters between the data items.  In Hugh Darwen’s book “An Introduction to Relational Database Theory,” he uses tables to explain the theory behind a relational database.  In a way, a table can be seen as a relational database, where the various parameters are presented in a way that lends itself to relation.  However, Darwen argues that, while the visualization is similar, a table is not really a relational database in itself.  In order to be a proper relational database, the columns of a table must be algebraically or operationally related to one another. [1]  This aspect of relational databases are the primary benefit to using them in the organization of big data, especially in the construction and building industry.  The ability to operate on parameters within a database based on known relations allows the user to construct a systematic collection of information that is pertinent in the building.

     To understand the effects that this type of information can have on our industry, the mathematical theory is useful to grasp what should and should not be related.  Eliezer Lozinski explains this theory rather well in his journal “Construction of Relations in Relational Databases.”  He explains that all databases are made up of relations that are made up of a collection of attributes.  These attributes are linked through a set of dependencies to the relation.  This framework for relating the attributes is called a scheme.  The scheme of a relation can be represented graphically, as Lozinski does below.  This rather simplified progression shows how the various schemes can be put together to relate the appropriate attributes.  [2]

Figure 1: Relational Scheme Tree [2]

     The impacts to the construction industry are numerous.  As we heard in the Whiting Turner guest lecture, a lot of firms are pushing to tabulate the entire building, from door hardware to reheat coils.  Relational Databases can take it a step further, by joining the parameters of installed components, we can understand the connections between key parameters and design, maintain, and modify building components better.

Comments on Others

Andrew Ng – Your explanation of Structured Query Language was very helpful, particularly with the examples that you gave.  This system would be very useful to a project manager in this way to cut back a lot on the unnecessary work that stems from handling big data. 

Prahlad Singh – I particularly liked your post because of all the benefits you provided to having databases in our industry.  I think that both designers and contractors are constantly overwhelmed with large amounts of data, and any process by which we can eliminate raw number sorting/crunching can make the entire process more efficient.

Ray Powell – Your post was really interesting to read since it connects what we have been learning so far to our next topic.  Both BIM and databases are essential for design firms now and it was really interesting to see how the two can interact to make the design process more efficient.

References

[1] Darwen, Hugh.  “An Introduction to Relational Database Theory.”  Darwen & Ventus Publishing ApS 2010.  Web.  Available: https://dvikan.no/ntnu-studentserver/kompendier/an-introduction-to-relational-database-theory.pdf

[2] Lozinski, Eliezer.  “Construction of Relations in Relational Databases.”  The Hebrew University, Jerusalem, Israel.  Web.  Available: http://delivery.acm.org/10.1145/330000/320155/p208-lozinskii.pdf?ip=129.25.36.43&id=320155&acc=ACTIVE%20SERVICE&key=A792924B58C015C1%2EE0B7266DB2A396BF%2E4D4702B0C3E38B35%2E4D4702B0C3E38B35&CFID=900189139&CFTOKEN=74586775&__acm__=1487008321_4d9c6ffee5fc5fe79479552f6b170344