Interacting Objects Process Model

We can apply the systems methods proposed on these pages to physical objects, such as building structures.

Conventially they are modelled as action and reaction over time. The loads acting in a gravitational field are the action potential that causes a flow of force through a structure which is then reacted by the foundations. More generally all physical systems consist of a potential (e.g. a voltage, water pressure head) driving a flow (current – amps, water flow rate) through impedances (capacitance, inductance and resistance, tanks, natural curves, boundary layer drag).

These processes can be simulated on a computer as a response to time-varying demands (potential). The Interacting Objects Process Model (IOPM) is different from conventional methods in that each element (e.g. a beam or column in a structure or each element in an electrical network or finite element in the analytical scheme) is modelled as an individual process interacting with its related processes which we call its friends.

In a conventional analysis the physical connections between the physical objects are modelled through the degrees of freedom (dofs) in matrix equations modeling the flows. In the IOPM, these dofs are effectively channels of communication linking friendly processes. These processes interact by sending messages (down these channels of communication), which are the current values of the state variables such as position, velocity, acceleration, force, etc. Each process (e.g. finite element) transforms an input message from friendly processes into output messages, which are sent back to friendly processes.

Each process acts in parallel. The IOPM is so-called because each process is modelled using a set of software objects in an object-oriented computer language such as C++. The results for standard problems are identical to traditional methods. The technique is versatile, flexible and simple to use. It allows complex transformations (e.g. non-linear ‘chaotic’ dynamics) to be modelled easily and quickly, and can be extended to solve problems well beyond the scope of conventional methods. For further details see  Blockley, 1995.