Scribd
Upload
43 views7 pages

Inventory Management Simulation Model

The document describes a simulation of an inventory management system for a company that sells a single product. It outlines the objectives to compare ordering policies and decide inventory levels. The model simulates events like demands, orders, and inventory evaluations over 120 months. It tests 9 inventory policies defined by reorder points (s) and order-up-to levels (S). The model tracks inventory levels, backlogs, costs to evaluate policies and inform decisions on how much inventory to keep each month.

Uploaded by

abeni mesfin
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
43 views7 pages

Inventory Management Simulation Model

The document describes a simulation of an inventory management system for a company that sells a single product. It outlines the objectives to compare ordering policies and decide inventory levels. The model simulates events like demands, orders, and inventory evaluations over 120 months. It tests 9 inventory policies defined by reorder points (s) and order-up-to levels (S). The model tracks inventory levels, backlogs, costs to evaluate policies and inform decisions on how much inventory to keep each month.

Uploaded by

abeni mesfin
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd

UNITY UNIVERSITY

DEPARTMENT OF COMPUTER SCIENCE

SIMULATION OF INVENTORY MANAGEMENT SYSTEM

Group members ID

1. Binyam Tesfaye UU70320R


2. Robel Addis UU70324R
3. Bereketab Tariku UU70327R
4. Amanuel Abiy UU70470R
5. Kaleab Tesfaye UU70244R
6. Leul Garaw UU71230R
7. Yidnekachew Tilahun UU71225R
SIMULATION OF AN INVENTORY SYSTEM
Problem formulation
A company that sells a single product would like to decide how many it should have in inventory for
each of the next n months. The times between demands are exponential random variables with a mean
of 0.1 month. The sizes of the demands, D, are random variables (independent of when the demands
occur), with

D = {(1,1/6), (2,1/3), (3,1/3), (4,1/6)} where x = demand and y = probability in (x, y)

Setting of objectives and overall project plan


- To compare alternative ordering policies for an inventory system.
- To decide how many items a company should have in its inventory.

Model conceptualization / Development


Our model of the inventory system uses the following types of events:

Event type Event description

1 Arrival of an order to the company from the


supplier
2 Demand for the product from the customer
3 End of the simulation after n months
4 Inventory evaluation (and possible ordering) at the
beginning of a month

At the beginning of each month, the company reviews the inventory level and decides how many items
to order from its supplier. If the company orders Z items, it incurs a cost of K + iZ.

Where K = 32 is the setup cost and I = 3 is the incremental cost per item ordered. (If Z = 0, no cost is
incurred). When an order is placed, the time required for it to arrive (called the delivery lag or lead
time) is a random variable that is distributed uniformly between 0.5 and 1 month.

The company uses a stationary (s, S) policy to decide how much to order,
Z = S -1 if I < S, 0 otherwise
Where I, is the inventory level at the beginning to the month.
When a demand occurs, it is satisfied immediately if the inventory level is at least as large as the demand.
If the demand exceeds the inventory level, the excess of demand over supply is backlogged and satisfied
by future deliveries. (In this case, the new inventory level is equal to the old inventory level minus the
demand size, resulting in a negative inventory level) When an order arrives, it is first used to eliminate as
much of the backlog (if any) as possible; the remainder of the order (if any) is added to the inventory.
Most real inventory systems also have two additional types of costs, holding and shortage costs, which
we discuss after introducing some additional notation. Let let) be the inventory level at time t
[note that let) could be positive, negative, or zero], let I+(t) = max{l(t), O} be the number of items
physically on hand in the inventory at time t [note that 1+ (t) ;" 0], and let ret) = max{ - l(t), O} be the
backlog at time t [r(t);" 0 as well. The holding cost includes such costs as warehouse rental, insurance, taxes,
and maintenance, as well as the opportunity cost of having capital tied up in inventory rather than
invested elsewhere. We have ignored in our formulation the fact that some holding costs are still incurred
when I+(t) = O. However, since our goal is to compare ordering policies, ignoring this factor, which after
all is independent of the policy used, will not affect our assessment of which policy is best.

Now, since 1+(t) is the number of items held in inventory at time t, the time-average
(per month) number of items held in inventory for the n-month period is

I+ =

The time-average number of items in backlog is

The demand-size random variate D must be discrete, as described above, and can be generated as
follows. First divide the unit interval into the contiguous subintervals C, = [0,1/6), C2 = [1/6,1/2), C3 =
[1/2,5/6), and C4 = [5/6, 1], and obtain a U (0, 1) random variate U from the random-number generator. If
U falls in C,' return D = 1; if U falls in C" return D = 2; and so on. Since the width of C, is i - 0 = i, and
since U is uniformly distributed over [0,1], the probability that U falls in C, (and thus that we return D = 1)
is i; this agrees with the desired probability that D = 1. Similarly, we return D = 2 if U falls in C2, having
probability equal to the width of C2, 1/2 – 1/6 = j, as desired, and so on for the other intervals. The
subprograms to generate the demand sizes all use this principle, and take as input the cutoff points
defining the above subintervals, which are the cumulative probabilities of the distribution of D.

The delivery lags are uniformly distributed, but not over the unit interval [0,1]. In general, we can
generate a random variate distributed uniformly over any interval [a, b] by generating a U(0,1) random
number U, and then returning a + U(b - a).

Data Collection
Assume that the initial inventory level is 1(0) =60 and that no order is outstanding. 'We simulate the
inventory system for n = 120 months and use the average total cost per month (which is the suni of the
average ordering cost per month, the average holding cost per month, and the average shortage cost per
month) to compare the following nine inventory policies:
s 20 20 20 20 40 40 40 60 60
S 40 60 80 100 60 80 100 80 100
The state variables for a simulation model of this inventory system are the inventory level l(t) , the
amount of an outstanding order from the company to the supplier, and the time of the last event which is
needed to compute the areas under the l\t) and r(t) functions.

Model Translation / Programming

Flowchart for demand routine, inventory model.

Demand
Event

Generate the size


of this demand
Decrement the
inventory level by
this demand size

Schedule the next


demand event

Return

Flowchart for inventory evaluation routine, inventory model.

Inventory
evaluation event

IS

YES I(t) < S NO

Determine amount
to be ordered

S = I(t)
Incur ordering
costs and gather
statistics

Schedule order
arrival event for
this order

Schedule order
arrival event for
this order

Return

Flowchart for routine to update the continuous time statistical accumulators, inventory model

Update time
average statistical
accumulators

Was I(t) in the


Negative previous positive
interval
positive,
negative or zero

Update area under

I-(t)
Zero
Update area under
I+(t)

Return

Common questions

Powered by AI

The simulation model uses discrete random variables to determine the sizes of customer demands. It divides the unit interval into contiguous subintervals corresponding to the probabilities of different demand sizes. A uniformly distributed random number U(0,1) is generated: if U falls in the interval [0,1/6), demand size D is set to 1; in [1/6,1/2), D is set to 2; in [1/2,5/6), D is set to 3; and in [5/6,1], D is set to 4. These intervals are determined based on the demand probabilities given as {(1,1/6), (2,1/3), (3,1/3), (4,1/6)} .

The key costs associated with managing inventory in the simulation include ordering costs (setup and incremental costs), holding costs (expenses related to storage such as warehouse rental, insurance, taxes, and maintenance), and shortage costs (costs due to backorders and potential lost sales). These costs are used to evaluate different inventory policies by calculating the average total cost per month over a simulated period of 120 months. The total cost per period includes the sum of average ordering, holding, and shortage costs, thus providing a comprehensive assessment of each policy’s efficiency .

Simulation plays a critical role in this study as it allows for the assessment of various inventory policies under uncertain and dynamic demand conditions without actual risk to operations. It provides detailed performance insights on costs, service levels, and inventory utilization by evaluating policy efficacy over an extended period of simulated months. This process helps identify optimal strategies aligning with goals such as minimizing total costs while maintaining adequate stock levels to meet demand .

The (s, S) policy in this inventory management simulation is a replenishment strategy where an order is placed only when the inventory level falls below a defined reorder point, s. When this occurs, an order is placed to raise the inventory level to a higher threshold, S. The policy functions by ensuring that an order sufficient to restore inventory from below s to S is made, minimizing both ordering and shortage costs while keeping inventory levels sufficient to meet incoming demand .

The model simulates lead times for order deliveries using a uniform distribution over the interval [0.5, 1] months, meaning that the time between placing an order and its delivery is equally likely to fall anywhere within this range. This choice assumes no extreme variability in delivery performance, suggesting a reliable and predictable supplier relationship where lead times are consistent without long delays that could disrupt inventory replenishment .

Ignoring certain holding costs such as those incurred when inventory is at levels different from zero simplifies the simulation model, reducing complexity and focusing computational resources on critical cost factors directly affected by policy changes. While this can lead to inaccuracies in total cost estimation, particularly if these fixed costs are significant, the approximation is deemed acceptable since the objective is to compare relative performance across ordering policies rather than exact cost accounting. The core insight into policy efficiency derives more from variable costs like ordering and shortage rather than fixed holding costs .

The random time between demands impacts the timing of inventory stock depletion and replenishment orders. Modeling it as an exponential random variable allows for a memoryless property where the time until the next demand is independent of previous demand intervals, capturing the natural variations in demand occurrences. This impacts how an inventory system must plan for fluctuations, accommodating both regular and sporadic demand patterns efficiently .

The decision-making process for placing an order involves evaluating the current inventory level at the beginning of each month. If the inventory level I is less than the reorder point S, a new order is placed for the quantity Z = S - I. Otherwise, no order is placed. The costs involved in placing an order include a setup cost of K = 32 and an incremental cost per item ordered I = 3. The delivery of the order, once placed, arrives after a random lag time uniformly distributed between 0.5 and 1 month .

Initial conditions such as the starting inventory level significantly influence the simulation outcomes. A higher starting inventory provides buffer against demand surges early on, potentially reducing initial shortage costs and affecting inventory policy effectiveness. Conversely, lower starting levels might increase the likelihood of immediate backlog or higher costs initially, skewing the performance perception of an inventory policy. These conditions set the baseline from which policies are measured, influencing how rapidly equilibrium between supply and demand is established during the simulation period .

Backlogging occurs when demand exceeds available inventory, temporarily allowing inventory levels to become negative. These unmet demands are tracked and fulfilled as soon as future orders arrive, prioritizing the reduction of backlog before restocking positive inventory levels. This mechanism ensures that high customer service levels are maintained even under fluctuating demand, although it may temporarily increase shortage-related costs and affect cash flows and operational flexibility .

You might also like