...implemented

This demonstrator is described in Appendix .

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...Model

The name of the model is given by the name of the fictitious company that is modeled. It will hereafter be referred to as the PMC Model

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...

Also referred to as customer service

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...successful

``The implementation of JIT also entails shifting the management of supply inventories onto the suppliers.'' Grossman et al.[14]

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...environment

The environment may be the physical world, a user via a graphical user interface, a collection of other agents, the Internet, or perhaps all of these combined

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...intentions

The reader must not confuse these intentions with those expressed in the :intent slot of the message.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...costly

Detailed discussion in Section .

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...agents

The agent names are written in bold font, with capitalized first letter.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...Statistics

These are shown as the thinnest arrows in Figure . Which conversation classes used for the reporting, is shown in Table .

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...graphs

In the case of the PMC Model, the graph name is precede by the name of the agent that recorded the data. For example in Figure  (page ) the data is recorded by WW-Transport-inc.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...peak

The lower peaks occuring two weeks before is related to consignment units, and not further discussed.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...A/F

Ratio of actual to forecasted demand. A ratio of 0.5 means that only 50% of what was forcasted came in as actual orders. A/F = 0.75 means that 75% of forecasted orders came in, A/F = 1 that 100% came in, and so forth. This is true for all three products.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...uncertainties

Exept for scrap rates in the workstations. These are set to a certain percent, i.e. each unit leaving a station has this percent change of being scrapped. This uncertainty has little impact on the results of the simulation.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...delivered

The average is weighted by product values, and are calculated each week for the orders delivered that week.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...levels

Each row of graphs represent one of the plants' RPI. From top to bottom: SYS-Plant, CBOX-Plant, and MB-Plant. Black lines gives actual inventory, gray lines calculated safety stocks.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...MB-A1)

Simulations were run with various parameters. These parameters were chosen to be presented because of the clearity of the results

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...avoided

Already ordered materials will make the curve climb somewhat.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...given

Note that these graphs are the same as are presented in fig .

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...100%

in week 74 for A/F=1.25 and week 40 for A/F=1.5

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...times

The reason for this is that certain assumptions must be made regarding the future sharing of common ressources in production. A more powerful algorithm would probably do a better job, but there is reason to believe that the uncertain future makes it impossible to quote 100% correctly.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.