Inventory Control Models

1. What factors influence the choice of an inventory

control models;
2. Selective inventory control models – case studies
3. Definition of terms used in inventory management
practice
4. Inventory Models for independent demand - how
often inventory levels are reviewed to determine when and
how much to order
5. Inventory Models for dependent demand
6. Inventory Models for multi-echelons levels
 Inventory Management
Deciding on
the right service level 1. How often the inventory status should be determined
for a certain product is (periodic or continuous review?)
2. When to place an order ( reorder point)
essentially balancing
3. How much to order (order quantity)
inventory costs vs. the
cost of a stock out.
System criteria Strategic

•Customer Structure of inventory •Cost

Service objectives
management system objectives

Tactical Inventory costs

- ordering
Demand Control variables - holding
Purchase Lead time - purchasing
Service level - shortage

• How should inventory be managed (point to reorder, how much to order, ..)?
• Why does inventory fluctuate and what strategies minimize this (uncertainty
demand , supply process or other)?
1.

= piesa de schimb
multicomponenta

 Inbound Logistics Handling Equipments

 Nature of Demand – Dependence of items
Tracking inventory level is an essential part of business operations for any company
that sells physical goods. But, depending on the nature of demand there is different
concepts, formulae, and methods of inventory management. We need to distinguish
between two basic inventory categories:

Independent Demand (uncertain)

Finished goods, items that are ready to be sold
A E.g. a computer

Dependent Demand (certain)

B(4) C(2)
Components of finished products
E.g. parts that make up the computer
D(2) E(1) D(3) F(2)

We need to distinguish between two basic inventory categories: independent

demand and dependent demand inventory.
The distinction between the two is crucial because the tools and techniques
needed to manage each are very different.
 Nature of Demand
Inventory Control Systems
INDEPENDENT DEMAND INVENTORY - the products are
independent of one another.
For independent demand is used Inventory Control Systems based
on reorder-point (ROP) formulae to determine how much of a given
item needs to be ordered. In this case a reorder point (ROP) is set for
each item. The ROP is the lowest amount of an item you will have on
hand and on order before you reorder.

• DEPENDENT DEMAND INVENTORY - is used Inventory Control

Systems based on MRP (materials requirements planning) - the
demand for raw materials, parts, and assemblies is dependent on the
demand for the final product. The objective of dependent demand
inventory control is to support the master production schedule which
sets out what will be built, when, and in what quantities.
Independent demand vs. Dependent demand
inventory management systems
 Inventory Control Policy

Key Questions:

What are the questions I should

ask to determine the type of
inventory control system to
use?

How important is the item?

–The more stock, the more How often inventory levels are
working capital is needed and reviewed? Should review be
the more stock depreciation you periodic or continuous?
get.
–On the other hand if you do When and how much to order?
not have enough stock, you
get inventory stock-outs, missing
What cost or service objectives
potential sales, possibility
should I set?
interrupting the whole production
process.
 Selective Inventory Management
cererea nu este distribuita in mod egal intre produse
a). Choosing categories
high service level, e.g. 96-98%, It is in our experience fully
sufficient to differentiate
between 3-5 service level
categories that cover the
product portfolio from must
medium service level, e.g. 91-95%
have items to the lowest
priority items. In this slide,
as an example, we chose a
three-value system:
lower service level, e.g. 85-90%
- High: 96-98%
- Medium: 91-95%
- Low: 85-90%

Remarks:
1. SKU stratification (A/B/C) is difficult to use in some business area (ex. a fashion business
use “get in, get out model”)
2. In practice, sales volume (turnover) is not the only metric that weighs the importance of an
item. Other rankings that are often used solely or in combination with turnover are profitability
(gross margin contribution), number of orders, COGS (cost of goods sold) and all should also
influence the inventory strategy.
 b). VED - Items Classification
VED: Vital, Essential & Desirable classification
VED classification is based on the criticality of the inventories.
• Vital items – Its shortage may cause havoc & stop the work in
organization. They are stocked adequately to ensure smooth
operation.
• Essential items - Here, reasonable risk can be taken. If not
available, the plant does not stop; but the efficiency of operations
is adversely affected due to expediting expenses. They should be
sufficiently stocked to ensure regular flow of work.
• Desirable items – Its non availability does not stop the work
because they can be easily purchased from the market as & when
needed. They may be stocked very low or not stocked.

Obs. Clasificarea VED a articolelor se aplica in general pentru piesele de schimb

 Categorizing products using a mix analysis method – ABC & VED
• It is useful in capital intensive industries, transport
industries, etc.
• VED analysis can be better used with ABC analysis in the
following pattern:
Category “V” items “E” items “D” items

“A” items A-V A-E A-D

Constant control & Moderate stocks No stocks
regular follow up
“B” items B-V B-E B-D
Moderate stocks Moderate stocks Low stocks

“C” items C-V C-E C-D

High stocks Moderate stocks Very low stocks

In acest caz tehnica ABC a fost folosita pentru clasificarea produselor in functie
de costurile unitare de achizitie (mai costisitoare, costistoare, economice-normale).

Obs. Uneori se aplica clasificarea HML (High, Mediu, Low)- derivata din metoda ABC,
dar care ia in calcul valoarea unitara a produsului si nu un volum periodic (anual).
c). FSN Analysis – Items clasification
• FSN: Fast moving, slow moving & non moving (dead stock)
• Classification is based on based on their consumption
pattern or rate of movement and is useful in controlling
obsolescence.
• Date of receipt or last date of issue, whichever is later, is taken
to determine the no. of months which have lapsed since the
last transaction.
• The items are usually grouped in periods of 12 months.
• It helps to avoid investments in non moving or slow items. It is
also useful in facilitating timely control.
Obs. Prin analiza FSN se determina nivelul transactional al unui
produs care va sta la baza reanalizarii ofertei comerciale
pentru o anumita perioada (un an).
Se foloseste in special pentru a decide delistarea produselor
din categoria non moving.
 Mix analysis method
quantity & frequency of sales
 2. Four Types of Inventory Control Systems
How often inventory levels are reviewed? Should review be periodic or continuous?
When and how much to order?

 Evaluarea continua - Continuous review Continuous Review

system or Fixed-order-quantity models (Q) -
Event-driven
Reorder policies that are driven by  (s, q) System - Order-Point (s), Order-
event time - inventory is continuously Quantity(q); if IP < s order q
monitored every day and when it drops to  (s, S) System - Order-Point (s), Order-
predetermined level s, an order of size Q is Up-to-Level(S); if IP < s order q=S-IP
placed .

 Evaluarea Periodica - Periodic review Periodic Review

system or Fixed-time-period models (R) –
Time Driven  (R, S) System - Order-Up-to-Level (S)
Reorder policies that are driven by
every R time periods order q = S-IP
event time - inventory is reviewed at
 (R, s, S) System - a combination of (s,
regular period intervals (R), and an order is
S) and (R, S) systems - every R time
placed to raise the inventory to a specified periods if IP < s order q=S-IP
level (order-up-to level,S).

S= stoc maxim; s= reorder point (nivelul de alarma); R – intervalul

de timp dintre 2 comenzi succesive; q- comanda de reaprovizionare
 Periodic
review –
Evaluarea
periodica a
stocului la
Continuu review – intervale
Evaluarea continua a fixe de timp
stocului
 Comparison of Periodic and Continuous Review Systems
How often inventory levels are reviewed? Should review be periodic or continuous?

Periodic Review Continuous Review

• Fixed order intervals (R) • Varying order intervals
• Variable order sizes • Fixed order sizes (q)
• Know inventory level only at certain times • Continuous collecting information about
inventory level to make replenishment
• Orders are released (lansate) periodically
decisions
according to supplier schedule agenda for
for multiple stock items • Very useful for types of individual items:
Higher-priced, critical, or important
• Ordering costs are lower
• Allows the customer to individually
• Possible quantity discounts to reduce the review the frequency of orders
number of small orders and orders a Pros/Cons
truckload once
• Replenishment made dynamically
Pros/Cons • Lower, less-expensive safety stocks for
• No need for a system to continuously the same service level
monitor item (IP) • Cost of handling equipments higher
• Convenient to administer -coordination of • Management of Inbound logistics flow
replenishment, unload &download planning harder
• Higher safety stocks • Each stock item reordered at different
times - complex, no economies of
• Higher carrying costs - not optimum, but
transport runs 15
more practical
 3. Terminology used in inventory model
Reorder Point with a Safety Stock
S – order-up- to Level
R – Review
q – order d - Demand Period (Order
quantity rate cycle)

Place order
intervalul de
timp între
două

Receive order
Q s aprovizionări
succesive
s – Reorder point

t t
Ss Purchase
Safety Stock Lead Time Time
0
Stoc siguranta
(timpul / perioada de (perioada de aprovizionare = R+t+
reaprovizionare) perioada de calcul pentru Ss)
Variable demand with Safety Stock to satisfy demand that exceeds the amount forecasted.
 Terminology and calculations
• Timpul/perioada de reaprovizionare - Purchase Lead time (t)
The purchase lead-time is the time from the ordering decision until the ordered amount is
available on shelf.
The purchase lead-time = order preparation time + transit time for the order+ administrative
time at the supplier+time for inspection after receiving the order.

• Ciclul comenzilor: Order Cycle Time (R) - perioada dintre 2 comenzi

succesive
In bussines practice are called replenishment order cycle, refers to the time between orders
of a specific item. Most easily calculated by dividing the order quantity by the annual demand
and multiplying by the number of days in the year.

• Perioada de prognozare a cererii: Forecast period

Perioada pentru care se face prognoza cererii este egala cu: Order Cycle Time (R)
perioada dintre 2 comenzi succesive + Purchase Lead Time (t) perioada de
reaprovizionare + perioada utilizata la calcularea stocului de siguranta.
Perioada de prognoză utilizată la calculul stocului de siguranță poate diferi de perioada de
prognoza a cererii. De exemplu, este posibil să aveți o perioadă pentru care se face
prognoza cererii de patru săptămâni, în timp ce perioada pe care o utilizați pentru calculul
stocului de siguranță sa fie de o săptămână.
 Notation and definitions of inventory level

• Inventory On-order (Inv on order) = the number of units that have been ordered but have not been
received (comenzi deschise catre furnizori sau COMANDA FURNIZOR IN DERULARE sau Stoc in Tranzit).

• inventory On-hand (IOH) - the number of units physically in inventory ready to serve demand (stocul
fizic disponibil pentru vanzare/utilizare: nu include stocul expirat, deteriorat, indisponibil pentru
vanzare/utilizare).

• Committed – comenzi transmise catre depozit care asteapta sa fie colectate pentru a fi livrate
clientilor(stoc rezervat)

• Backorders - comenzi neonorate din lipsa de stoc - the total amount of demand that has not been
satisfied due to a stock out:
– All backordered demand is eventually filled, i.e., there are no lost sales.
• Inventory level = Inventory On-hand (IOH) – Backorder - Committed.

• Inventory position (IP) = Inventory On-order + Inventory On-hand (IOH) – Backorder - Committed.
• IP – stocul disponibil calculat
• s=Order Point- Nivelul stocului la care trebuie sa se declanseze actiunea de reaprovizionare (nivelul de
alarma); S=Order-up-to Level: nivelul maxim planificat al stocului; q – Order Quantity - comanda de
aprovizionare; R- Review Period - perioada dintre 2 comenzi succesive de reaprovizionare; t- Purchase
Lead Time – perioada (timpul) de reaprovizionare
 Models for Inventory Management
 When and how much to order?
 What cost or service objectives should
Inventory level depends essentially of two key factors:
– purchase lead time (t)
– demand (Q)

Both two factors are subject to uncertainties

- purchase lead time variations: suppliers or transporters may
be faced with unplanned difficulties
- demand variations: customer behaviors can evolve in rather
unpredictable ways.

Safety stock (SS): stock that is held in excess of expected

demand due to variable demand rate and/or variable lead
time.
 Factors Affecting Inventory Policy Decisions
Market Characteristics Demand (cererea)
Demand may be:
– deterministic, the demand for an item is known exactly and in advance
– constant, the demand for an item remains the same from one period to
another
– variable, the demand is not constant and varies with time, not known but
demand characteristics such as mean, standard deviation and the
distribution of demand may be known by using forecasting technique
when historical data are available.
– stochastic/probabilistic, the demand is not known, but based on historical
data we may be able to fit it to a probability distribution
Uncertainty in demand forecasting and the degree of collaboration or
information sharing are tightly linked. The greater the degree of collaboration,
the lower the coefficient of variation of the demand model.
See in the next slides how can you effectively manage your variable demand.
4. Classification of inventory models
 Timpul R intre 2
comenzi succesive = 0
(Evaluarea continuă)
q size fix q size variabil

s= la ce nivel cantitativ al stocului se

EOQ model – in cazul cererii lanseaza comanda de reaprovizionare
constante s = (d*t)+ Ss; s=stoc minim

Continuous Review model is very useful for article types: Higher-priced, critical, or important
Nu este util in cazul aprovizionarii produselor care au sortimente multiple – comenzi separate pt
fiecare sortiment de paine
S= stoc maxim; s= reorder point (nivelul de alarma); t = Purchase Lead Time - timpul
de reaprovizionare (timpul dintre momentul lansarii comenzii si momentul receptionarii
acesteia; d – cererea medie zilnica; Ss – stocul de siguranta
Continuous Review, Order-Up-to-Level (s, S) System
A Simple Two - bin Inventory System*
Other type of
Bin 1 Bin 2 Inventory System
Min-Max System is:

• Special case of fixed

order quantity model.

• Amount of stock
equivalent to the reorder
point is physically
separated into a second
Bin 2.
MIN, MAX sunt valori
setate in sistem.
Q=MAX-MIN
*This assumes that lead time is exact, there are no vendor stock-outs or backorders,
and that there are never any defects. That assumption is, of course, often FALSE.
 IKEA - Maximum/Minimum Settings
As Proprietary System Inventory Control

The in-store logistics managers use an inventory replenishment management

process developed by IKEA called ‘minimum/maximum settings’ to respond to
store-level inventory reorder points and reorder products.

Minimum settings: The minimum amount of products available before

reordering.
Maximum settings: The maximum amount of a particular product to order
at one time.

Using IKEA’s proprietary inventory system, logistics managers know what is

sold through point-of-sale (POS) data and how much inventory comes into the
store through direct shipping and from distribution centres through warehouse
management system data.
From these data, they can forecast sales for the next couple of days and order
in the suitable amount of products to meet that demand.
 Deterministics Demand Inventory model
Economic Order Quantity (EOQ)

Economic Order Quantity (EOQ) - model de gestiune cu cerere constantă,

perioadă fixă de reaprovizionare a stocului, fără permisiunea rupturii de stoc.
EOQ use continue review method with fixed-order-quantity systems (Q);
Need to order EOQ at the Reorder Point

EOQ Restrictions:
• Demand for the product is known & constant
• Lead time is known & constant (fixed)
• Purchase price per unit is constant (no quantity discounts)
• Instantaneous receipt of material
• The ordering (setup) cost is constant (not depending on the quantity ordered)
• Use only inventory carrying cost dependent on the average annual inventory size
• All demands for the product will be satisfied (no back orders or stock outs)
 Continue review method with Q size fix
Economic Order Quantity (EOQ)
• EOQ or Fixed Order Quantity system is the technique of ordering materials whenever stock reaches
the reorder point.
• Economic order quantity deals when the cost of procurement
(annual ordering cost) and handling of inventory (annual holding cost) are at optimum level and total
cost is minimum.

Average inventory = stocul mediu =

= q/2 +Ss, atunci când modelul
lucrează cu stoc de siguranță

Annual Annual Annual

Total Annual Cost* (TC)
Purchase + Holding + Ordering
(costurile totale de gestiune) = Cost Cost
Cost
TC = Total annual cost($); D = Demand per year (pcs)
C = Purchasing cost per unit, is fixed ($);
q = Opimal order quantity (pcs)
cl = Cost of purchasing order/setup cost ($)
cs = Cost of holding a single unit in inventory for a year ($)
q/2 = stocul mediu anual
 EOQ Model Equations

D =Demand per year (pcs)

Based on reorder point - d = Demand per day (pcs)
When inventory IP is < s,
order replenishment of q, t = Purchase lead time in days
meaning EOQ. s = Reorder point
Cs = Cost of holding a single unit in
inventory for a year ($)* - costul de stocare
anual al unei unitati din produs in depozit.
Cl = Cost of purchasing order/setup cost
($)
 Economic Order Quantity - EOQ MODEL
Tc (Total Cost)

Annual Cost ($)

Holding/Carrying Cost

Ordering Cost

EOQ
Order Quantity (Pcs)
Optimal Order Quantity (q)
The trade-off of holding and ordering costs
The objective is to decide the optimal (economic) order quantity q per order so
as to minimize the total annual cost, including holding and ordering costs.
As you can see in the graph above, the optimal order quantity (q) is achieved at the
point where yearly holding costs equal yearly ordering costs:

Obs. Deoarece pretul de achizitie (purchase price) al produsului comandat este constant
(nu variaza in functie de cantitatea comandata), acesta nu este utilizat in modelul EOQ.
 Study Case: EOQ Model

Find the economic order quantity and the reorder point, given
• Annual demand ( D ) = 1,000 units
• Average daily demand* (d ) = 1,000/365 sau 1000/250
• Ordering cost ( Cl) = $5 per order
• Holding cost ( Cs ) = $1.25 per unit per year
• Purchase lead time ( t ) = 5 days
• Cost per unit ( C ) = $12.50

What quantity should be ordered?

Average daily demand* (d ) – cererea medie zilnica

 EOQ Model Solution
SOLUTION:
 s - la ce nivel cantitativ al
stocului se lanseaza
(Evaluarea periodică) comanda de reaprovizionare
s = (d*t)+ Ss

q = S - IP S=d(R+t)+Ss q = S – IP numai daca IP<=s

d= cererea medie zilnica; R – intervalul de timp dintre 2 comenzi succesive (ciclul comenzilor);
t- timpul de reaprovizionare; R+t= perioada pentru care se efectueaza aprovizionarea;
d – cererea medie zilnica; Ss – stocul de siguranta; S-stocul maxim/tinta; IP- Inventory Position/Stoc disponibil
calculat; q- comanda de reaprovizionare
 Deciziile privind momentul si cantitatea de
reaprovizionat in modelul de stoc Periodic review
(evaluare periodica a stocurilor)
Order point (s) - punctul de comanda al reaprovizionarii- s=(L*d)+ Ss si reprezinta
nivelul stocul cantitativ (bucati) la momentul deciziei de lansare a unei comenzi de
reaprovizionare.
Explicatii:
a) s – punctul de comanda al reaprovizionarii – in practica logistica se mai
numeste si stocul minim.
b) t – Purchase lead time -timpul de reaprovizionare (timpul dintre momentul
lansarii comenzii si momentul receptionarii acesteia)
c) d – average demand -vanzarea medie zilnica pe o perioada istorica stabilita a
vanzarilor. Pentru acuratete se recomanda calcularea vanzarii medii zilnice prin
impartirea vanzarii totale din perioada stabilita la numarul de zile pentru care
a existat stoc (pentru zilele fara stoc nu au existat vanzari).
d) Ss – stocul de siguranta
Stocul maxim/tinta (S)- nivelul tinta al stocului- S=(R+t)*d)+Ss, unde R este Review
period, perioada dintre 2 comenzi succesive de reaprovizionare; L- Purchase Lead
Time - timpul de reaprovizionare.
Order size (q) - dimensiunea lotului de reaprovizionat, respectiv comanda de
aprovizionare – q=S-IP, IP – Inventory Postion – stoc disponibil calculat :
IP = Inventory On-order + Inventory On-hand (IOH) – Backorder - Committed.
 Periodic Review Systems
Calculations for ”S” (stoc maxim) and ”q” (Qty Order)
• Targeted Inventory level (such as Max level):
S = d(R + t) + Ss (use Ss due to uncertainty demand)
d = average period demand
R = review period (days, wks)
t = Purchase lead time (days, wks)
Ss = zσL, unde σL= standard deviation (deviatia standard) of demand during lead
time (which must be calculated in order to find safety stock)
Aceasta formula de calcul a stocului de siguranta Ss considera ca,
cererea in perioada de reaprovizionare (t) urmeaza o distributie
normala (Gauss).
z= safety factor corresponding cycle service level during order lead
time (L) – see next slides

• Replenishment Quantity (q)=S-IP > 0

» q= order size
» S= target inventory
» IP= Inventory position
 Study case:
Modelul de Evaluare Periodică: determinarea stocului tinta
(S) si a comenzii de aprovizionare (q)
Fie urmatoarele informatii referitoare la importul unui produs :
• stocul de siguranta Ss= 328 unitati (pentru simplificare este deja
determinat).
• vanzarea medie zilnica a produsului pe o perioada de 6 luni:
D = 1000 unitati/luna
• stocul disponibil calculat IP = 1200 unitati ; IP= comenzi deschise + stocul
disponibil pentru vanzare – stocul rezervat – comenzi neonorate
• R: 3 luni - perioada de timp dintre 2 comenzi este de 3 luni
• t: 1 luna - perioada de reaprovizionare a stocului de la furnizori este 1 luna
Se calculeaza:
a) Nivelul stocului tinta/necesar:
S = D × (R + t) + SS = 1.000 × (3 + 1) + 328 = 4.328 buc.
a) b) Necesarul de aprovizionare: la fiecare 3 luni, compania examineaza
stocul si plaseaza o comanda de reaprovizionare q:
q =S-IP = 4.328 - 1.200 = 3.128 buc..
In ABC analysis, which items would use Periodic Review - system?
 SAFETY STOCK ANALYSIS

When both lead time & demand are What Happens when either Demand or
constant, you know exactly what the Lead Time Varies?
reorder point is: Avem nevoie de un stoc suplimentar (de
Reorder point = demand during lead time siguranta) pentru a satisface cererea
R=d * t peste medie din timpul perioadei de
aprovizionare (t): R= + Ss
 The Role of Safety Inventory in a Supply Chain
• Forecasts are never completely accurate
– If average demand is 1000 units per week, every once in a while actual demand is 1000. But about half
the time actual demand will be greater than 1000, and about half the time actual demand will be less
than 1000
– If you kept only enough inventory in stock to satisfy average demand, half the time you would run out
Safety Stock CONCEPTS:
 Safety Stock:
Buffer added to on hand inventory (stocul disponobil pentru vanzare) during lead
time for the purpose of satisfying demand that exceeds the amount forecasted in
a given period.
 Stockout - an inventory shortage
The cost of stock outs are much more complicated to determine.
A customer that does not find the product in store might either choose an alternative
that is in store, postpone the purchase to a later date or buy at the competition.
In grocery retail for example, out-of-shelf situations of certain must have products are
known to drive customers out of the store, taking their business to a competitor.

 Safety stock reduces risk of stockout during lead time

 Cycle Service level

 probability that the inventory available during lead time will meet demand

Conclusions: Very Important - Safety stock and Cycle Service level refers to lead
time.
 Managing Uncertainty in the
Supply Chain: Service level
The balance inventory costs vs. stock-outs costs is very business dependent. Thus, instead of considering those
costs directly, we will now introduce the difference between customer service level and cycle service
level.
Cycle-service level: The desired probability of not running out of stock in any one purchase lead
time cycle, which begins at the time an order is placed and ends when it arrives. The service level is
the probability that there are no backorders or stockouts during the replenishment lead
time.
Naturally, when safety stocks are increased, the service level increases as well.
When safety stocks get very large, the service level tends toward 100% (i.e. zero probability of encountering stock-out or 1 to
meet demand).

By example, if a retail has a cycle service levele of 95%, this means, 0.95 is probability to meet demand or 0.05 to have stock-
out during purchase lead time.

Customer Service Level vs. Cycle Service Level

• Cycle service level is concerned with our ability to meet demand during the order lead time (L),
when we are most in danger of running out
• In contrast, Customer Sevice Level (such as fill rate) refers to the percent of demand that we are
able to meet from stock across all time periods.

Rule of Thumb:
The resulting fill rate for a product will be higher than the cycle service level.
 Different methods of calculating the safety stock (Ss –
stocul de siguranță) in control systems under
probabilistic demand
• Equal Lead Time Supplies
Perioada utilizata la calcularea stocului de siguranta (Ss)
pentru largi categorii de produse (dacă nu toate) din
nomenclatorul articolelor aflate pe stoc este egala cu n
perioade de aprovizionare, unde n=1, …, n.
• Percentage of lead-time demand
By ex. SET safety stock at 50 percent of the lead-time
demand;
• Equal Safety Factors (k) – see next slides
Ss= kσL, unde k este factorul de siguranta si este acelasi
pentru o gamă largă de elemente.
• Computing Safety Stock (SS) using standard
deviation of demand during lead time

Ss = z σL
 L  t L
σt = known standard deviation over some time interval t
σL = standard deviation of demand during lead time
(which must be calculated in order to find safety
stock)
z = safety factor corresponding cycle service level during order
lead time (L) – see next slides
 Demand variation analysis using a normal
distribution

Historical Analysis of Demand History Normal Distribution

Interpretare statistică: cererea reala se afla in intervalul +/- 1 sigma in proportie de

aproximativ 68%, in +/- 2 sigma aprox. 95.5% si in +/- 3 sigma in aprox. 99,73% din
cazuri
 Different Forecast and
Lead Time Intervals
• The standard deviation is measured over the lead time for safety stock
calculation
• Many items in inventory have different lead times
• Sales records and forecasts are usually made on a weekly or monthly interval
so the lead time and the forecast interval i.e. (sales interval) are
different
• The standard deviation will change as the interval changes
– the longer the interval the higher the variation
• Need to adjust the standard deviation for the lead time

 LTI = FI * Lead Time Interval Example: FI= 4 weeks, LTI=2 weeks,
Forecast Interval FI =150 units.
Calculate the standard deviation for 
LTI
where : *LTI = Lead Time Interval σLTI = 150 * 2/4 = 150 x 0.707 =
*FI = Forecast Interval
106 units
 Exemplu de calcul al punctului de
reaprovizionare (s-Order point)

vanzarea medie zilnica: d = 18, timpul de reaprovizionare: L (lead time) =

2 week, t = 1 week, t = 5 (standard deviation/week)

L = t x L=5x 2 = 7.1

Safety stock = zL = 1.28 (7.1) = 9.1 or 9 units

Reorder point = L*d + Safety stock

= 2 (18) + 9 = 45 units

Concluzia: Cand nivelul stocului atinge 45 de unitati se initiaza o

comanda de aprovizionare.
Study case: Safety Stock calculation
– Here is a list of z values for different values of the cycle service level
α. Example: for α = 95% cycle service level (5% stockout during
lead time) , z= 1.65
Service Level and the Safety Factor, z
Service 90% 91% 92% 93% 94% 95% 96% 97% 98% 99% 99.9%
Level α
z 1.29 1.34 1.41 1.48 1.56 1.65 1.75 1.88 2.05 2.33 3.08
Statistic:
Cererea reala se afla in intervalul +/- 1 sigma in proportie de aproximativ 68%,
in +/- 2 sigma aprox. 98% si in +/- 3 sigma in aprox. 99,88% din cazuri
Fill Rate vs. Cycle Service Level
Cycle service level is concerned with our ability to meet demand during the order
lead time (L), when we are most in danger of running out
In contrast, fill rate refers to the percent of demand that we are able to meet from
stock across all time periods.

Rule of Thumb:
The resulting fill rate for a product will be higher than the cycle service level.
The “right” cycle service level for a given fill rate can be found through experience.
 5. Inventory Systems for Dependent Demand
• Dependent Demand – component parts of finished products,
subassemblies, raw materials, etc- used to produce final products.
• Dependent demand is certain and is dependent on the demand
for finished products.
Inventory Systems for Dependent Demand:
– MPS (Master production schedule )
– MRP (Material Requirements Planning ) -Planificarea necesarului de
materiale. Planned order release ignores the capacity of the manufacturing
facility or the supplier to produce or supply that order quantity in the time
frame required. This may make MRP process infeasible. To circumvent this,
MRP must be integrated with capacity requirement planning (CRP).
– MRP II (Manufacturing Resources Planning) = MRP+CRP (capacity
requirement planning) - Planificarea necesarului de materiale si a
cerintelor capacitatilor de productie

– MRP/JIT
 MRP Evolution
Open-Loop Schedule Materials
MRP
Closed Loop Schedule Materials
Incorporate Feedback from the production activities
MRP
Schedule & Purchase Materials
MRP II Coordinate workflow/ Mfg Resources
Units are PUSHED forward according to
the plan!

Units are PULLED forward only

MRP/JIT when needed!

ERP
 Flow chart of a MRP process
Comenzile
Comenzile previzonate
ferme de la Forecasts
Firm orders Aggregate
clienti of demand
from known product from random
customers plan customers

Master
production
schedule
Engineering
(MPS) Inventory
design
transactions
changes
Material
Bill of Requirements Inventory
material planning record
File (BOM) (MRP) file
Reports
Production Structure System 12
Material Requirements Planning (MRP)
• Based on primary inputs as MPS, BOM’s
(for part relationships) and Inventory
Records (for lead times and inventory
position) a MRP system:
– Creates schedules identifying the specific
parts and materials required to produce
end items
– Determines exact numbers needed
– Determines the dates when orders for
those materials should be released, based
on lead times
11
 Master production schedule (MPS)
MPS - the schedule of what will be built, when
(time), and in what quantities (pcs).

An example of MPS:
Month October November December
Week 41 42 43 44 45 46 47 48 49 50 51 52

PC 300 300 300 300 300 300 300 300 250 250 250 250
Product

Notebook 300 300 300 300 250 250

Tablet 300 300 300 300 250 250

Totals 2,400 2,400 2,000

 An example of Bill of Material (BOM)*
A Complete Product Description
Process Lead Time
(10)
A A bill of material (BOM)
defines the product structure for a
manufactured item.

Question:
When do we start
B(2) (15) C(1) (10) producing/ordering each part for
50 PCs?
See MRP to create a schedule
to satisfy demand.

(15)

(10) (15)
D(3) E(3) D(1)

*A Bill of Material (BOM) defines the product structure (raw materials and
subassemblies) for a manufactured item.
 Flow chart of MRP II

MRP does not provide a link

with other functional areas of
business such as finance and
marketing.

This may lead to non-

coordinated functioning of
production and procurement
system.

In order to further refine the

process, MRP II
(manufacturing resources
planning) and ERP
(enterprise resource
planning) have been
developed to take planning
process in a more integrative
concept at the enterprise
level.
6. Multi-echelon supply chain
 Multi-echelon supply chain - example
Multi-echelon supply chain
System is composed of stages (nodes,
sites, …). Each node/stage has exactly
one predecessor. 2. Two- echelon retail supply chain
Stages are grouped into echelons network
• Stages can represent: Physical
locations, Processing activities , etc The chain network contains two
intermediate nodes to the end customer
1. Three-echelon repair network
(aggregated)
Supplier 1 Supplier 2 Supplier 3
Supplier Supplier
lead time lead time
Echelon

Warehouse 1 Warehouse 2 Echelon 2

Central depot
Distributor
lead time
Intermediate depots
Retailer 1 Retailer 2 Retailer 3 Echelon 1

Operating sites Transshipment – transfer intern de marfa

Remarks: In a distribution system each node/stage is referred to as echelon 53

and the number of echelons is critical for inventory policy.
 Managing inventory in a distribution company
with multi-echelon supply chain
• A single-echelon supply chain model - a single retailer, group of
retailer, or stores do not ship to each other. They occupy a single level in the
hierarchy. Ex: Supplier-Retailer-Customer
• A multi-echelon supply chain model - upstream locations
having finite quantities delivered across large numbers of
downstream locations.
Ex: Supplier-Warehouse-Retailer-Customer – two echelon supply
chain network
2 assumptions:
1. Inventory decisions are made by a single decision maker whose
objective is to minimize system wide cost
2. The decision maker has access to inventory information at each of the
echelon node/stage
Under these assumptions, an inventory policy based on the so-called echelon
inventory is an effective way to manage the system.
Factors Affecting Inventory Policy Decisions
• Multi - echelon structure
• The central warehouse, regional
warehouses, branches, and technician
levels in the diagram each represent
an echelon in the system in our service
parts system example.
• Inventory policies as well as
information flows need to be
coordinated throughout a supply
chain. Their echelon structures are
sometimes affected by governmental
requirements for local content, tax
policies, labor rules (legislatia muncii),
cost, etc.
 Managing inventory in the multi echelon inventory Supply Chain
Case study: Two-Echelon Distribution - they occupy two level in the hierarchy.
A two-echelon-distribution inventory system consists of a single warehouse
and multiple retail locations.
The echelon inventory at any stage or level of the system is equal to the inventory
on hand at the echelon, plus all downstream inventory.
IOH (Warehouse echelon inventory) =
the inventory at the warehouse + all of the inventory in transit to and in stock at the
retailers – la stocul din esalonul superior se aduna toate stocurile din esaloanele
inf.;
IP (Warehouse echelon inventory position) = IOH (the echelon inventory at the
warehouse) + Inventory On-order (those items ordered by the warehouse that have
not yet received - all items that are backordered - committed .

Exemplu:
Le Purchase lead time Warehouse – Supplier: 7 zile
Purchase Lead time IOH
Warehouse – Retailers: 5
zile
Lead time Warehouse – Retailers:
Purchase Lead time 5 zile Source:
Warehouse – Supplier: 7 David Simchi-Levi
zile

Le = 7+5=12 zile
 Effective approach to managing inventory in the Supply Chain
of distribution company using (s, S) inventory policy

The individual retailers and warehouse

ordering decisions are managed using the
appropriate (s, S) inventory policy
• For each retailer , the reorder point, s,
and order-up-to level, S, are calculated
 Whenever the inventory position
at a retailer falls below the reorder
point s, an order S-IP is placed to
raise its inventory position to S S-
stoc before order, S- stoc
• For the warehouse, the reorder point s maxim
and an order-up-to level S is calculated
for the warehouse
 Whenever the inventory position
for the warehouse is below s, and
order S-IP is placed to raise the
echelon inventory position
to S
 Managing inventory in the SC
Reorder point calculation (nivelul
stocului care determina lansarea comenzii
de reaprovizionarea)

s  Le * AVG  z * STD Le
where

– Le = Purchase echelon lead time – perioada de reaprovizionare ( zile

/saptamani)
The lead time between the retailers and the warehouse plus the lead time
between the warehouse and its supplier
– AVG = average demand across all retailers (i.e., the average of the aggregate
demand) – cererea medie agregata pe (zile/saptamani) – cererea totala la
nivelul esalonului superior=suma cererilor esaloanelor inferioare
– Ss = z*STD√Le – stocul de siguranta
– STD = standard deviation of (aggregate) demand across all retailers
– z = safety factor
 Case study: Network Topology

Retailer 1
• Distribution system:
Supplier Distributor Retailer 2

Retailer 3

1. Average weekly demand faced by the retailer is 45

2. Standard deviation of demand is 32 Retailer 4
3. At each stage, management is attempting to maintain
a cycle service level α of 97% (z=1.88)
4. Purchase lead time between each of the stages
(retailers, distributor) and between the distributor and its
suppliers is 1 week .
• Safety Stock calculation
– Here is a list of service factor (z) values for different
values of the cycle service level α. Example: for α = 95%
service cycle level (5% stockout during lead time) , z=
1.65. Use Excel function NORMSINV to convert service
level percentage to service factor.

Cycle service Level and the Safety Factor, z

Cycle 90% 91% 92% 93% 94% 95% 96% 97% 98% 99% 99.9%
Service
Level α
Service 1.29 1.34 1.41 1.48 1.56 1.65 1.75 1.88 2.05 2.33 3.08
Factor z

Statistic
1. Cererea reala se afla in intervalull +/- 1 sigma in proportie de aproximativ
68%, in +/- 2 sigma aprox. 98% si in +/- 3 sigma in aprox. 99,88% din cazuri
 Reorder Points at Each Stage

• For the retailer, s=1*45+1.88*32*√1 = 105

• For the distributor, s=2*45+1.88*32*√2 = 175
• For the supplier, s=3*45+1.88*32*√3 = 239
where reorder point (s) is calculated by:
s  L * AVG  z * STD L
e e

Use Excel function STDEVPA to calculate standard

deviation.
 Planificarea aprovizionarii: Modele de prognoza
Selectie model in functie de tipul cererii

Tip Model Test de verificare

T: Test pentru trend

Model Constant
Model cu tendinta S: Test pentru sezonier
Model sezonier A: Test pentru trend si
Model Sezonier cu tendinta sezonalitate

Modele Parametrii

Valorile de baza

Valorile de baza
Valorile de trend

Valorile de baza
Indicii de sezonalitate
 Planificare aprovizionare: Overview

Articole

Prognoza
indicator
?

Analiza lista
cerinte

Articol cod 10
Vanzarile Prognoza cererii Articol cod 20
lunare
Cerinte
10/2005 987
11/2005 1000 aprovizionare
12/2005 1251
Comenzi aprovizionare.

Articol cod 10
Articol cod 20
 Aplicatia de Prognoza
Determina cerintele de aprovizionare

Planificarea
Reorder punctului
Point
de reaprovizionare
Planning Planificarea dimensiunii
La ce volum al stocului Lotului de aprovizionat
lansez o noua comanda?

s=Order Point Valorile prognozate

reprezinta cerintele de
aprovizionare
Calculeaza punctul de
reaprovizionare si eventual
Analiza cerintelor rezultate
volumul stocului de
pentru validarea comenzilor
siguranta.
de aprovizionare.