Árbók VFÍ/TFÍ - 01.01.1997, Blaðsíða 246

244 Árbók VFÍ/TFÍ 1995/96
Buyers Graph
200 •
180 •
160 •
140 ■
120-
100 •
80-
60-
40-
20-
0-1----»— ----«------«----->------1------1 ____.______._____,_______,
0 2000 4000 6000 8000 10000 12000 14000 16000 18000 20000
AddlUooal Raw Material bought [kg]
Fig.3
point B on Fig. 1, which for a certain range of the ratio R/M results in an optimal combination
of net profit contribution per unit of each of the two bottlenecks. Because of this, and because
a concept like duality is not very well known in the industry, Fig.l is called the “Contribu-
tion Graph”.
A prototype of a graphical system as is described here was developed by the authors using
Excel 5.0 and Visual Basic. The system begins by presenting a table for editing the basic
assumptions, see Table 1, and after that Fig. 1 appears. When the user selects with the mouse
a point on Fig. 1, for example A, B or C, the system calculates the corresponding product mix
(or simply the quantity to produce in case of a single product solution) and the total net profit
contribution.
If the user wishes, he can bring up Fig. 2, i.e. the graph of the primal model for the two
products in the solution, showing the two bottleneck resources as constraints. Therefore, we
have called Fig. 2 the “Bottleneck Graph”. Also the objective function is shown on the graph.
Furthermore, the user can order a so-called “Buyer’s Graph”, see Fig. 3, which is a
shadow price graph showing the shadow price yR, that is the maximum attractive price to pay
for additional units of the raw material, as a function of increasing R. This could become an
important decision aid when buying fish on the daily fish markets, possibly in the future on a
lap top computer. A similar graph for the shadow price of manpower, which increases as yR
decreases, could be added as an aid for decisions regarding overtime work.
All these graphs, and the assumption table, are stored in the Excel 5.0 system in one
workbook, making it very easy to jump between sheets. The system has been demonstrated
for several production managers in Icelandic fish processing plants and the responses have
been very encouraging, although further dialog and requirements analysis is necessary. Also,
more programming effort is needed to integrate the system into the software environment in
freezing plants and potentially also into software on-board of factory trawlers.
s
£