Exercise Two: schematic design |
The Coopers, an energetic, professional, middle-income family of three (parents and a 16 years old girl), desires to build their own house, based on the GIMS Building System. The house will have two floors, with the traditional separation of the private spaces from the semi-public ones according to floors (bedrooms on the second floor). The family is quite hospitable, so a guest bedroom is desired. 
Case 1. The weight of the adjacency is smaller
Case 2. The weight of the adjacency is greater Case 3. The weight of the adjacency is the smallest |
Floor Plan 2.
 
Floor Plan 3. In my experience in using GIMS, computers can assist people in generating design solutions to some degree. This software can evaluate a scheme by comparing the "weights" that we put into the "Adjacency Matrix" boxes and the "distances" between spaces allocated by designers, and then give us a "value" to tell us how "good" is the scheme. The smaller the value, the better the design scheme. Thus it helps us improve our floor plan step by step. But as architects, we should not totally depend on this software. Any software has its shortcomings, the GIMS software only provides objective solutions that sometimes do not help us make correct judgments. For instance, the value of case 2 is much greater than that of the case 1, but we cannot see any difference between the two floor plans, except that the "weight" we assign to each of the "Adjacency Matrix" is quite different. So we cannot tell if a scheme is better only from the "value" of the layout. We must figure out what causes the discrepancy of "value" and reevaluate them ourselves. For the same reason, we cannot say that case 3 is better than case 2 only because the "value" of the former is smaller than that of the latter. The floor plan of case 3 seems so unreasonable. Because the "weight" of the adjacency between "entrance" and "kitchen" equals that between "entrance" and "master bedroom", so the system allocates the space randomly, thus it puts "master bedroom" on the 1st floor as well as the "kitchen" on the 2nd floor.
In my point of view, as an experimental software, it should be improved in following aspects: 1. Enlarge the number of rooms other than mandatory ones (now we can only add to the mandatory spaces up to TWO more rooms, for some family, it is much less than enough); 2. Enlarge the choices of different combination of rooms (there are so many different family structures that people need more choices to suit their life-style); 3. Permit people to choose the number of floor other than TWO; 4. Provide flexible sizes of rooms (now the room size is constant--14’X14’); 5. Provide some storing space, and some uncertain space to meet the various needs of people.
Here are three successive designs shown that the value of the layout has been improving (the left ones are proposed by the system and the right ones are my own space allocations): |
