Sunteți pe pagina 1din 10

Selecting the right inventory target strategy for minimizing

inventory levels for products with high variation in the forecast


error using Monte Carlo simulation
Jeroen Marc Roger Vrouenraetsa
03-04-2013
a Faculty
of Technology, Policy and Management, Delft university of Technology, Jaffalaan 5, Delft, The
Netherlands

Article info Abstract


Keywords: In supply chain management, analytical inventory models are used to
prescribe a target safety stock to achieve optimal inventory levels.
Inventory management, These models prescribe either a fixed safety stock or dynamic safety
Monte Carlo simulation, time. In the existing literature little attention is given to the impacts of
supply chain management, either settings on the inventory performance based on the analytical
safety time, safety stock inventory models. In this paper is determined: What is the optimal
target safety setting strategy for products with a high variation in the
forecast errors using distribution resource planning, considering
inventory size and service levels? First, a literature review is discussed
on the little known effects on the use of either setting. Based on a case
study, a simulation model is developed to determine the optimal
target setting. The results of the simulation study quantify the
negative effects when using safety time for products which have a high
variation in forecast errors. Furthermore a valid inventory simulation
model for experimenting purposes is proposed.

1 Introduction To achieve target service levels, literature


describes the use of analytical models. These
The fast moving consumer goods (FMCG) sector models describe reality with mathematical
is one in which having an optimal supply chain equations. They attempt to capture the
(SC) is a great competitive benefit. The goal in stochastic nature of inventory drivers in order to
each SC as described by Goldratt & Cox (2004 ), calculate a minimum inventory buffer which
is to (1) maximize the rate at which the system covers for uncertainties in the SC (Silver &
generates money through sales, while (2) Peterson, 1985).
minimizing the money that the system has
invested in purchasing things that it intends to There are two type of strategies to set the target
sell and (3) minimize the money that is spend in minimum inventory buffer: fixed safety stock or
order to turn inventory into throughput. dynamic safety time. A fixed safety stock reflects
a minimum number of products to keep in stock.
In line with this goal, inventory management is The safety time targets inventory replenishment
pursuing to achieve target service levels (1) by a certain period before the actual need is
balancing inventory (2), supply chain expected(van Goor, Kruijtzer, & Esmeijer,
capabilities and demand (3). Selected measures Goederenstroombesturing, voorraadbeheer en
of performance are the service level and the materials handling, 1990).
average inventory size.

Scientific article – JMR Vrouenraets 1


Although both strategies can be calculated using 2. Literature review
the same analytical equations, it is unclear what
benefits and risks are related to which strategy. 2.1 Inventory replenishments cycle
In the literature little has been written on the
Schneider (1981) summarizes several
different effects of the different strategies on the
definitions for the service level of a supply chain.
SC performance.
A convenient measure to express service is the
The objective of this paper therefore is to fill the so-called case fill rate (CFR), which is the
gaps in literature by adding knowledge on the fraction of demand which can be directly
effects on performance measures in the SC satisfied from the shelf.
because of the use of different stock setting
A way to improve the CFR is to keep an optimal
strategies. The following question is central in
amount of safety stock. This is a part of the
this paper: What is the optimal target safety
inventory that is held in excess of expected
setting strategy for products with a high variation
demand due to variable demand rate and/or
in the forecast errors using distribution resource
lead time (Stevenson, 2005). Once a target safety
planning, considering inventory size and service
stock level is determined for the SC using
levels?
accurate data, the setting would not have to
This paper will give examples on the use of the change until high impact changes in the SC take
different stock setting strategies through a case place (e.g. extra trade lane which significantly
study conducted within an international FMCG shorten the lead time).
SC.
Once a target is set, one should focus on
By means of combining a literature review on maintaining the stock levels near the target
the use of the setting strategies, the use of an safety stock level by means of replenishing
inventory Monte Carlo simulation model for optimal quantities of stock, called cycle stock.
testing several scenarios and the experiences Multiple methods are available for calculating
during the case study, the effects are analyzed. the right amount of cycle stock, which depends
on economical and technical constraints(van
The contribution of this paper is the Goor, Kruijtzer, & Esmeijer, 1990). E.g. the lot for
quantification of the negative effects when using lot strategy replenishes the minimum quantity
safety time for products which have a high which is necessary to align with the target safety
variation in forecast errors. Furthermore a valid stock and technical constraints, but this strategy
inventory simulation model for experimenting does not consider economical constraints like
purposes is proposed. ordering cost(Silver, Pyke, & Peterson, 1998).

The structure of this paper is as follows. Section A system to control the time-phased
2 provides a literature review to define the requirements of replenishment for distribution
inventory management concepts necessary to centers is called distribution requirements
understand the importance of inventory safety planning (DRP) (van Goor, van Amstel, & van
settings. Also, considerations on the use of the Amstel, 1989). By monitoring the inventory
two safety setting strategies will be given. levels, (forecasted) demand and replenishment
Furthermore the concept and limitations of lead times, it suggest moments to replenish so to
analytical inventory models are discussed. keep inventory around the target safety stock.
Section 3 discusses the use of a Monte Carlo
simulation model for inventory management Wrong safety settings fail to cover for demand
after which section 4 describes the development due to unexpected variation during the lead
of such a simulation model based on a case study time, resulting in lower than targeted CFR.
and explores the use of different safety setting Hence, for the ongoing replenishment cycle, the
strategies. Section 5 states the conclusions and accurate planning of the initial safety setting is
answers the research question. Section 6 crucial. Widely used are analytical inventory
provides a discussion on the performed models to calculate an optimal safety setting
research. with.

Scientific article – JMR Vrouenraets 2


2.2 Analytical inventory models = Target service
…………….level
Analytical models describe a system using a set Equation 2 Q = Average order
of multiple equations. Analytical equations or …………….quantity per
…………….cycle
numerical algorithms are used to find one, point- = special
estimate solution for a problem (Haugh, 2004). …………….function of
…………….the standard
The analytical inventory model is used to …………….normal
calculate the target safety settings with, …………….distribution
developed by Silver & Peterson (1985). Others,
such as de Kok et al (2012), have improved this
A short reasoning; if a higher CFR target is
model by eliminating certain constraints of the
selected or the standard deviation of the
model.
estimated errors increases, the function
Equation 1 shows the key elements in the model returns a higher value for k and thus the safety
to calculate the safety stock with. factor is relative high.

SS = safety stock (product) On the other hand, if the order quantity


Equation 1 K = risk factor (unit less) increases, the safety factor decreases. This might
= Standard deviation of the seem counterintuitive but with a higher order
estimated error during the
replenishment lead time quantity, the cycle time increases and thus the
(product) number of replenishments in a period decreases
(ceteris paribus). Given the fact that the risk of a
stock out is higher near the end of a cycle (when
(1) The target safety stock. This should be used the stock is low), this risk now is reduced due to
as a target when the inventory is replenished so the fewer replenishments and thus a smaller
that a certain target CFR is achieved. safety factor can be used.

(2) The standard deviation of the estimated With a unit less risk factor and a standard
errors resembles the variation in demand and deviation expressed in number of products, the
supply chain elements during the replenishment safety stock setting resembles the number of
lead time. Little variations in these elements products that should be kept as safety inventory
make a supply chain relative predictable (in a to cover for the expected variation in demand
utopia, deterministic), thus a reason to keep less and lead time in order to achieve a target case fill
safety stock. rate. This is called the safety stock and is
constant unless reviewed.
The errors are driven by multiple elements in
the SC such as forecasting or production issues. By dividing the safety stock by the average daily
Silver et al (1998) propose an accurate method demand, a safety time (expressed in days) is
to combine these elements. calculated. Instead of using a fixed safety stock to
trigger replenishment, safety time sets the target
(3) The k-factor, or the risk factor, determines stock equal to the expected (forecasted) demand
how many times the standard deviation should over the safety time. This results in replenishing
be kept in stock in order to achieve a target CFR . inventory an amount of days before it is actual
needed, creating a safety time buffer(van Goor,
The value for k is driven by three parameters van Amstel, & van Amstel, 1989).
(Equation 2) and a special function of the unit
normal distribution which eliminates the 2.3 Differences between safety strategies
cumulative density for k smaller than 0 (Rosen,
2013). Literature on the use of safety time is limited, let
alone the differences in performance between
both safety strategies. While the conventional
analytical model simply suggest safety stock to

Scientific article – JMR Vrouenraets 3


be used as optimal safety setting, the first However, the effects of choosing between safety
question is why even bother to use safety time? stock and safety time are not clear. As stated by
Aguilar et al (1999), for such cases simulation is
General reasoning on safety time returns that it an effective tool to communicate results and
fluctuates with the expected demand. Thus, in performance dynamics.
periods of low demand, inventory benefits from
safety time because it reduces total inventory, Monte Carlo (MC) simulation is based on two
whereas, in times of high demand, it provides mathematical theorems which make it a very
extra security. useful simulation type to analyze inventory
behavior with. (1)the law of large numbers and
Rosen (2013) favors this reasoning, suggesting (2) the central limit theorem. Not only can MC
that safety time is useful near the end of a simulation, considering the first theorem, show
product life cycle is preferable. While evidently an estimate of the expected result, it also returns
the demand will decrease, the safety time will an estimate of the uncertainty in this estimate
make the total inventory decrease with the same (Dunn & Shultis, 2011). These characteristics
pace, resulting with little non-performing make MC useful to account for risk in
inventory at the end of the life cycle. quantitative analysis (Palisade, 2013).

Whybark and Williams (1976) pose that Recent studies have been using Monte Carlo
uncertainty in timing of demands should be simulation for inventory management problems
dealt with using safety time, whereas (Cáceres-Cruz, Grasman, Bektas, & Faulin,
uncertainty in quantity should be dealt with 2012)(Jaio & Du, 2010). However little effort has
using safety stock. been put specifically in using MC simulation to
explore the uncertainty involved with the
Chang (1985) argues that safety stock and safety
analytical model and the use of different safety
time are interchangeable. However, his modeling
setting strategies.
of production and demand in essence is
deterministic which makes is less valuable to Key differences between the use of an analytical
apply in practice. Yano (1987) focused on model and the MC simulation are shown in Table
finding the optimal planning lead time but also 1.
only considered deterministic demand.
Analytical MC
Buzacott and Shanthikumar (1994) with the use model simulation
of stochastic modeling that safety time is Input para- Static Used to draw
preferred over safety stock given the condition meters parameters to samples from
that forecast are accurate. Moreover, with describe
changing customer orders during the lead time stochastic
behavior with
or bad forecasts, inventory performance would
System Analytical Defines
result from fixed safety stock. characterist equations operating
ics behavior and
Although is general safety stock seems a robust functional
choice to optimize supply chain performance relationships
with, no studies have been found to quantify the Results Point estimates Range
impact. estimates
Table 1: Difference between analytical models and MC
3. Monte Carlo simulation simulation models

Simulation models mimics the operating


behavior of a system (Verbraeck & Valentin,
2006). To understand the most important
behavior and to create a simulation model often
is more time consuming than the use of an
analytical model.

Scientific article – JMR Vrouenraets 4


4. Case study show that for both the variation in forecast
errors and lead time errors a normal
In order to develop a Monte Carlo simulation distribution can be used. The simulation model
model to simulate inventory management has the property to set the average variation for
processes with, key system characteristics have the distribution. This allows for experimenting
to be selected. In order to do so a system with accurate and useless forecast and likewise,
analysis is performed at an FMCG supply chain. high lead time variations.
For the design approach of the simulation model Multiple experts with different expertise within
three viewpoints have been combined. First, the the SC, were interviewed for the validation of the
simulation method of Banks (1999) for the simulation model (Sarikaya, 2013)(van der Oost,
general modeling methods. Secondly, the META 2013)(Alves, 2013). Each expert is asked to
model (Herder & Stikkelman, 2004) is used. The check the dynamics by looking at the specific
model starts with defining the model parts of the model and the output graphs. All
requirements and the possible solution space to acknowledged the validity of the model. Alves
answer to these requirements with. Last, the recognized the fact that no inventory is
spiral model (Boehm, 1988) is used for the simulated at the plant but agreed to leave it out
approach. The spiral model consists of several of the simulation while it concerns a single stage
iterative stages in which the model is designed, simulation model. Moreover, using a binary
tested and adapted resulting in a robust model search, the average optimal safety stock setting
design. (after a 1000 iterations) for a 99% case fill rate
according to the simulation model was
To obtain requirements for the model, the TIP-
compared with the suggested safety stock
framework as proposed by Koppenjan and
setting of the analytical model. The range did not
Groenenwegen (2005) is used to analyze
show any significant deviation from the
inventory management from a technology,
analytical model which echoes the validity of the
institutional and process point of view.
simulation model.
The selected performance measures for the
Two experimental designs are conducted. For
strategies are (1) the case fill rate (%), (2) and
both experiments, 4 product classes (table 3)
the average inventory (Statistical units and
were defined using the criteria of daily volume
days).
(threshold; 100 SU/day) and COVFE (threshold;
The conceptual model consists of four parts: (1) 100%). Products from the company were
the sampling demand and a adhering forecast of selected. All supply chain parameters are
demand, (2) the daily inventory position, (3) the selected in such a way that it would reflect
DRP planning of order replenishments and (4) reality.
the production and actual replenishments of
For one experiment, the analytical optimal safety
products. These four parts together are able to
setting was used expressed as safety stock. For
simulate the behavior of inventory over time. Of
the other experiment, safety time was used.
key interest is the use of safety settings as a
trigger for the DRP planning to replenish. After using the method for calculating the
number of iterations as described by Verbraeck
For the specification of the model, all data could
and Valentin(2006), both experiments were
be retrieved from data sources located within
executed a 1000 times.
the company.
The results were analyzed both on the level of 1
However, for the sampling of stochastic values
iteration and on the MC simulation as a whole.
for the demand, forecast and lead times, accurate
Now, the results for the product class with high
probability distributions need to be selected.
daily volume and high COVFE are discussed.
Statistical test on sample data showed that a
gamma distribution fits the distribution of The analysis of 1 iteration using safety time
demand (Vrouenraets, 2013). Silver et al (1998) (Graph 1) show four dynamics:

Scientific article – JMR Vrouenraets 5


(1) The first peak in the target safety stock (1) Compared with both of the previous
(4) is exactly matched mainly because of graph of the inventory results, it
the good forecasting. The inventory goes immediately becomes clear the
up to match the demand and after the inventory with the use of safety stock is
peak in demand, it lowers again. more stable: Peaks are less extreme and
only need a shorter time to restore
(2) The creation of yellow NPI (1) increases around the inventory target.
rapidly when only safety time is used.
E.g. due to the last extreme over (2) With the use of fixed safety stock, also
forecast, a lot of yellow NPI is created. there is creation of some yellow NPI.
The duration of the yellow NPI wave is The largest peak happens near (2),
larger because it is replenished already driven by the over forecast of demand.
a few days earlier compared with the Given the fact that in all three scenarios,
previous scenario. The peak of yellow an over forecast of demand leads to
NPI is higher because the target safety yellow NPI, means that yellow NPI can`t
setting goes up as it ‘looks’ further in the be prevented with high COVFE.
future.
Although the inventory position also reaches
(3) One of the assumed benefits of safety zero in the scenario with the fixed safety stock,
time is that it should lower inventory in the final score on average total inventory and
case no demand is expected. On the one case fill rate, show better numbers compared
hand, this is true. With low forecast of with safety time.
demand, the target safety setting gets
Both iteration already explain some of the
near 0 which saves inventory costs (3a).
dynamic behavior, however results of the MC
However, at this point the difference
simulation tell even more on the risk profile of
between forecast and actual demand is
both safety settings.
rather low. Point (3b) gives an example
where no demand is forecasted which Graph 4 and Graph 5 show the ranges of the
takes the inventory to 0, but performance measures after a 1000 iterations. It
unfortunately gets surprised by a peak clearly shows how both the average inventort
in demand (1, in graph 3). Now, no stock and CFR score are better when safety stock is
can satisfy the unexpected demand used. Moreover, the ranges wherein these value
during the lead time: and this is exactly lie are more accurate with the use of safety
the type of demand safety stock should be stock, in other words, the extremes are less
used for. Demand hurts the most, when likely to be expected.
it is least expected. And that is why
there should be safety stock, which now 5. Discussion of results
is not the case.
Table 1 compares the ranking of the average CFR
(4) An example of where the safety time score for each product class for the safety
does prove its benefits is the moment settings scenarios. It clearly shows that in case of
when the forecast of demand is accurate s.k.u. with a high COVFE (Class I and III), a fixed
(2 in graph 3). The first weeks (4) the safety stock setting results in the best CFR score.
inventory goes up and down with the This also counts for class II s.k.u. Only, for the
accurate safety setting which provides class IV s.k.u. the current situation slightly
safety. outperforms the fixed stock strategy.

The analysis of 1 iteration using safety stock Not only does the use of fixed safety stock out-
(Graph 2), using the same initial conditions perform other strategies on the average CFR
(Graph 3) as with the previous iteration, show score, it also makes the scoring range more
other dynamics. Most important findings from narrow (not shown in table below). This means
this graph on the dynamics are:

Scientific article – JMR Vrouenraets 6


that less (extreme) outliers are expected when uncertainty involved with the analytical model
fixed safety stock is used. and the use of different safety setting strategies.

Table 2 compares the ranking of the average By developing and validating a simulation model
inventory levels for each class for the safety for inventory management, the possibility was
setting scenarios. Also for the inventory, all s.k.u. created to test the effects of both settings and
classes benefit most from a fixed stock setting. quantify the differences on performance
measures such as inventory and case fill rate.
Not only does the use of fixed safety stock out-
perform other strategies on the average The conclusion is clear. For DRP system the use
inventory level, it also makes the inventory of safety stock for products with a high variance
range more narrow (not shown in table below). of the forecast error, safety stock is preferred
This means that less (extreme) outliers are over safety time.
expected when fixed safety stock is used.
The development of a valid simulation model for
6. Conclusion and further research inventory management can be used for many
purposes. The focus of this research was to point
The use of safety time has positive effects when out the difference between two types of safety
the forecast is accurate. Inventory benefits from setting for products with a high variation in
accurate low forecasts, which result a decrease forecast errors. However, by changing initial
of the total inventory whereas high CFR score parameters in the model, different values for
are obtained in case of accurate forecasted peaks demand, the forecast errors or lead time errors
in demand. can be simulated. Moreover, single iteration
results can be used for educating purpose to
However, the negative effects of safety time
make clear the effects of both using wrong safety
perhaps are bigger than currently known at P&G.
settings values and the use of wrong safety
On the one hand, when over forecast are made,
strategies.
the dynamics of safety time enlarges the creation
of excessive inventory. On the other hand, when 7. Reflection
the forecast of demand is too low, there is not
enough safety stock to cover for the forecast The design of the simulation model has been
error. This leads to the reasoning why inventory performed with care. Moreover, the final
is now at its most vulnerable: Demand hurts the simulation model has been validated both by the
most, when it is least expected. And that is use of experts validation and a quantitative
exactly when (and why) there should be safety validation using the results of the analytical
stock. This is not the case with the use of safety model. However, still some distinct choices have
time. been made that can influence the results.

Safety stock is a robust approach for covering for First, the choice for the use of a lot for lot
errors during the replenishment lead time. replenishment strategy was driven by the case
Extreme low values for CFR scores or high study and the objective to minimize inventory.
values for total inventory, are not common like However, many supply chains are driven by cost
with the use of safety time. Moreover, the of the inventory as well, therefore likely to use
average scores for both CFR and inventory are other replenishment strategies. Adaption of the
better than when safety time is used according model then is needed to experiment for
to the simulation study. analyzing those effects.

Before this paper, little quantitative research Second, the use selection of sued probability
had been performed aiming at quantifying the distribution can affect the dynamics in the
differences between the use of two commonly results. Especially the gamma distribution has
used safety settings: safety stock and safety time. the feature of simulating a long tail, which might
Moreover, little effort had been put specifically not be accurate for certain products. This would
in using MC simulation to explore the make the current simulation model results more

Scientific article – JMR Vrouenraets 7


extreme than would be the case for the product Haugh, M. (2004, September). Overview of Monte
types. Carlo Simulation, Probability Review and
Introduction to Matlab. New York, USA.

Bibliography Herder, P. M., & Stikkelman, R. M. (2004).


Aguilar, M., Pater, A. J., & Rautert, T. (1999). Methanol-based industrial cluster design: A study
Business Process Simulation: A fundamental step of design options and the design process. Industrial
supporting process centered management. Winter & Engineering Chemistry Research , 3879 - 3885.
Simulation Conference (pp. 1383 - 1392).
Jaio, S., & Du, S. (2010). Modeling for Random
Amsterdam: Andersen Consulting.
Inventory System Based on Monte Carlo Theory
Alves, E. (2013, 01 25). DRP planning at P&G. (J. and Its Simulation. Third International Symposium
Vrouenraets, Interviewer) on Information Science and Engineering (pp. 296 -
399). IEEE.
Alves, E. (2013, March 18). Validation of simulation
model. (V. Jeroen, Interviewer) Koppenjan, J., & Groenewegen, J. (2005).
Institutional design for complex technological
Banks, J. (1999). Introduction to simulation. Winter systems. International Journal of Technology,
simulation Conference (pp. 7 -13). Atlanta, USA: Policy and Management .
AutoS.
Palisade. (2013, February 07). About Monte Carlo
Boehm, B. W. (1988, May). A spiral model of Simulation. Retrieved February 07, 2013, from
software development and enhancement. www.palisade.com:
Computer Vol 21, Issue 5 , pp. 61-72. http://www.palisade.com/downloads/factsfax/Wh
atIsMonteCarloSimulation.pdf
Buzacott, J., & Shanthikumar, J. (1994, December).
Safety Stock versus safety time in MRP Controlled Rosen, O. (2013, January 29). Dr. Ir. (J.
Production systems. Management Science Vol 40 , Vrouenraets, Interviewer)
pp. 1678 - 1689.
Sarikaya, E. (2013, January 21). Ir. (J. Vrouenraets,
Cáceres-Cruz, J., Grasman, S. E., Bektas, T., & Interviewer)
Faulin, J. (2012). Combining Monte Carlo
Simulation with heuristics for solving the inventory Schneider, H. (1981). Effect of service-levels on
routing problem with stochastic demands. Winter order-points or order-levels in inventory models.
Simulation Conference (pp. 1 - 9). Barcelona: Open International Journal of Production Research , 615-
University of Catalonia. 631.

Chang, C. A. (1985). The interchangeability of Silver, E. A., & Peterson, R. (1985). Decision
safety stocks and safety lead time. Journal of systems for inventory management and production
operations management , 35-42. planning. John Wiley & Sons Inc.

de Kok, A., Fortuin, L., & van Donselaar, K. (2012). Silver, E. A., Pyke, D. F., & Peterson, R. (1998).
Analysis of stock control models for one location Inventory management and production planning
with one product. Eindhoven, Noord-Brabant, The and scheduling. New York: John Wiley & Sons.
Netherlands: Technische Universiteit Eindhoven.
Stadtler, H. (2004). Supply chain management and
Dunn, W. L., & Shultis, K. J. (2011). Exploring Monte advanced planning––basics, overview and
Carlo Methods. Elsevier. challenges. European Journal of Operational
Research 163 , 575–588.
Goldratt, E. M., & Cox, J. (2004 ). The Goal: A
process of ongoing improvement. North River Stevenson, W. J. (2005). Operations Management.
Press. The McGraw-Hill Companies Inc.

Scientific article – JMR Vrouenraets 8


van der Oost, M. (2013, March 26). Validation of Vrouenraets, J. (2013). A step forward to
the simulation model. (V. Jeroen, Interviewer) optimizing inventory management. Delft: TU Delft.

van Goor, A., Kruijtzer, A., & Esmeijer, G. (1990). Whybark, D., & Williams, J. (1976). Materials
Goederenstroombesturing, voorraadbeheer en Requirement Planning under uncertainty. Decision
materials handling. Leiden/Antwerpen: Stenfert Science (7) , pp. 595-606.
Kroese Uitgevers.
Yano, C. (1987). Setting Planning Leadtimes in
van Goor, A., van Amstel, M. P., & van Amstel, W. Serial production Systems with earliness Costs.
P. (1989). Fysieke distributie: denken in Management Science , pp. 95-106.
toegevoegde waarde. Leiden/Antwerpen: Stenfert
Kroese Uitgevers.

Verbraeck, A., & Valentin, E. (2006). Discrete


modellen Deel II Distcrete simulatie. Discrete
modellen Deel II Distcrete simulatie . Delft, Zuid-
Holland, Nederland: Faculteit Techniek, Bestuur en
Management.

(4) (3b) (3a)

(1)

(1)

(2)
Graph 1 Results of iteration with use of safety time

(1) (2)

Graph 2 Results of iteration with use of safety stock

(2) (1)

Graph 3 Input for demand and forecast of demand for both iterations shown above

Scientific article – JMR Vrouenraets 9


Graph 4: Results average inventory score of Monte Carlo simulation using safety time and safety stock

Graph 5: Results of CFR score of Monte Carlo simulation using safety time and safety stock

Average CFR ranking Class I Class II Class III Class IV


Scenario 1 (safety time) 2 (80,6%) 2 (87,1%) 2 (92,5%) 2 (98,3%)
Scenario 2 (fixed stock) 1 (95,8%) 1 (97,8%) 1 (99,0%) 1 (99,9%)
Table 1: Ranking of scenarios on the average CFR score per s.k.u. class

Average inventory ranking Class I Class II Class III Class IV


Scenario 1 (safety time) 2 (6928 SU) 2 (1304 SU) 2 (1768 SU) 2 (44 SU)
Scenario 2 (fixed stock) 1 (3640 SU) 1 (985 SU) 1 (705 SU) 1 (41 SU)
Table 2: Ranking of scenarios on the average inventory level per s.k.u. class

High COVFE (group A) Low COVFE (group B)


High average demand (Group 1) Class I Class II
Low average demand (Group 2) Class III Class IV
Table 3: Recap of s.k.u. classification

Scientific article – JMR Vrouenraets 10

S-ar putea să vă placă și