The area of manufacturing economics lies at the
interface between engineering and business, and
has offered us the opportunity to solve problems
critical to not only the operation of manufacturing
processes, but also the viability and profitability
of the manufacturing business and the industry
in general. There are a host of issues within
this area, but we have thus far concentrated on
supply chain design and management and the economic
analysis of manufacturing business opportunities.
In the economic analysis of business opportunities,
we employ several tools, including two that we'll
discuss a bit further here. Cost models are developed
most often as a comparative tool, used to evaluate
alternative manufacturing processes or materials.
These are generally spreadsheet-based, and the
results can be used in other models that incorporate
the dynamics of the system under investigation.
The dynamic model is the other tool that we have
used extensively, and has been used in two forms:
the systems dynamics variety, and the discrete
events variety.
With the adoption of GoldSim, we have begun to
meld these tools together, using the capabilities
of the software to incorporate discrete events
in a system dynamics model, and to read and write
variable values to/from the spreadsheet in a dynamic
fashion.
Our focus is on performing economic analyses
and production economics research within the automotive
industry. We work with a number of clients, including:
Alcan Aluminum, General Motors of Canada, Transport
Canada and a number of Tier 1 and 2 suppliers
to the industry.
In the past, a dynamic cost analysis would begin
with a static process-based cost model (in a spreadsheet
format) of product manufacturing, and the outputs
would be utilized as inputs to a dynamic simulation
package. Now with GoldSim, we can utilize the
static model as both a database and to carry out
calculations as the simulation is running. In
addition, as we alter assumptions in GoldSim,
we can export those to Excel and update the spreadsheet
automatically.
We've developed an in-house generic template
for process-based cost modeling, which allows
us to quickly "fill in the blanks" for
almost any process, and was designed to work with
dynamic inputs from GoldSim. Once we have that
static model, it requires very little resources
to develop a GoldSim model to capture discrete
events and dynamic effects (e.g. inventory levels,
capacity increases, and equipment failures), giving
us a more detailed picture of the range of costs
associated with a particular product.
GoldSim Cost Model Spreadsheet - Screenshots
and Spreadsheet Layout ã2002 James McLeod
and the Centre for Automotive Materials and Manufacturing
Benefits of Utilizing Dynamic Exchange
There are a number of distinct benefits of dynamically
exchanging data between GoldSim and spreadsheets,
including:
1. It ensures that the system remains robust as the
model is refined, since assumptions/calculations only
have to be changed in one place, as they are defined
only once, in either Excel or in GoldSim.
2. It is easy to run multiple scenarios. Dynamic exchange
allows for more accurate (and effortless) examination
of the effects of changes in model assumptions, without
forcing one to update the spreadsheet for each set
of realizations, or rebuild simple calculations within
GoldSim.
Example Model - Polymer Component Plant In a current project, we are using both static
and dynamic approaches to direct product design and
development for a polymer component. Using GoldSim,
we have modeled the key components of the manufacturing
plant. Flow through the plant is controlled using
an order entry module, which takes market data from
the spreadsheet, and compensates for lost production
due to maintenance and equipment failure. The process
itself is reduced to three main stages: the compounder,
which mixes the raw ingredients; press equipment for
forming the part; and shipping.
Features 1. As the simulation progresses through time,
GoldSim's spreadsheet exchange mechanism allows us
to update raw material order quantities, as well as
reflect the changes in associated material cost with
time and as order quantities increase.
2. Presses can be added dynamically to the line when
required as product volume increases, and to replace
capacity lost to downtime.
3. Monitoring of scrap and inventory throughout the
model allows us to capture the variation in "work
in process" cost.
Results
Costs established in Excel are combined with the dynamic
costs calculated in GoldSim to provide a graph of
product cost versus time.
Please note that all values have been multiplied
by random factors such that the values on the graph
bear no relevance to the actual process. Currently
the model is deterministic, so the assumptions about
price and market size are fixed, but one can see the
natural variation in cost associated with a dynamic
system, even a deterministic one. GoldSim's stochastic
features will soon be incorporated so that we can
perform sensitivity analyses on variables like raw
material cost and market size.
The results obtained from this deterministic and
future stochastic models of the process will be used
to assist our client with determining if the level
of risk involved in pursuing this particular strategy
is offset by potential rewards. Concurrently, it will
provide decision support for determining research
and product development strategies.
