Uncertain demand: Normal distribution

• Normal Distribution: To relate service level and how many units to have in safety stock
• Bigger = better

meandemand un unitspertimeyear 1
unitspertimeyear I

z o

• Pasos:
• Data: µ, σ, x >/</=
1. Calcular z

z XF entre 2 valorescalcularporseparado
2. Buscar valor de z en tabla
3. Calcular z real
◦

IIEEEIIIIII
◦
◦ EII.EE
pro amor
rajen
Demand during lead time
• Place order when inventory reach R
• Order arrive after L periods
• Period of exposure to risk of stock out = L
• Normally distributed: When demand is independent across periods
• Bigger safety stock = bigger holding cost = bigger avoiding stockouts

meandemandduringleadtime must no L
standarddeviationofdemandduringleadtime Gust
Es
E z OLD
T
Prob. Of stock out during lead time
R Must Es OLDT

EOQ
24µF
Standard Normal Tables (Service Level)
serviceLeves SL z

Reorder Point and Safety Stock

SafetyStock SS Esa Out
R mio Es
Gust Must 55
Cost Function

Holdingcost h SS
totalvariablecost TVC holding setupcost
setup orderingcost
TotalAnnual cost TAC holding setupcost ad
Penaltycost pffff II
FEEF Fife
Periodic Review (T, S) Model
Info and formulas
• For "C" items
• Inventory position = checked at fixed intervals of time T
• Order-up-to level (S) = Maximum inventory level
• Order quantity = S - ending inventory at the time of review
• Service Level = Probability (demand during L + T) ≤ S
2 decisions and estimation fo S

Average inventory

m 2,500
6 500
L 2weeks
Review inventoryevery 4weeks
SL 90 1 7 1 28

FE 11.28145001 Es 1,567.67units
SS z O

S 15,000 12811,224 74 16,567 667

 Ssa z Gus 1.2817071067 905.096
Giro Op M SOOK 707.106
i You need move ss in periodic

ÉSE
S
L 2days
SL 95 z 1.64
s
S mar Z Out
Matt no Ath 18130 27 576
OLA OD EL 452 22 627
S 576 7.6422.627 613 1 614
Q 613 25 588bottles

Impact of product availability and uncertainity on safety stock

The appropriate level of safety stock depends upon…

• Uncertainty of both demand and supply: evaluated cases when lead-time is fixed and uncertain.
• The desired level of product availability (service level and fill rate)

Why must we take variability into consideration?

• The supply chains are breaking: electronics, fashion…
• Unexpected events…
• Transform traditional models… e.g., digitalization…

Key points: uncertainity models

• (Q, R) Policies:
◦ Both fill rate and service level increases as the safety stock is increased.
◦ For the same safety stock, an increase in lot size increases the fill rate, but not the service
level.
◦ The safety stock increases as leadtime variability increases.
◦ The required safety stock grows rapidly with an increase in the desired product availability
(Z values increase with an increase of SL).
• (T, S) Policies:
◦ With a continuous review policy, the safety stock is used to cover for demand uncertainty
over the leadtime.
◦ With a periodic review policy, the safety stock is used to cover demand uncertainty over the
leadtime and the review interval (L+T).
 ◦ Given than higher uncertainty must be accounted for, periodic review policies require a
higher level of safety stock.

Estimating and managing safety stock in practice

• First:
◦ Adjust inventory policies if demand is seasonal.
◦ Use simulation to test inventory policies.
◦ Start with a pilot.
◦ Monitor Service Levels.
◦ Focus on reducing Safety Stocks.
• Second:
◦ Reduce the supplier leadtime (and the variability of leadtimes):
‣ If the leadtime decreases by a factor of k, the required safety stock decreases by a factor
of sqrt(k).
‣ Strategies: faster transportation (air shipped), better coordination (information exchange,
retailer demand information).
◦ Reduce the underlying uncertainty of demand (represented by sigma(d)):
‣ If sigma(d) is reduced by a factor of k, the required safety stock also decreases by a
factor of k.
‣ Strategies: contracts, better forecasting, collaborative forecasting.
◦ A switch from periodic monitoring to continuous monitoring may also help reduce
inventories.
◦ Postponement (e.g., postponing product differentiation)

Materials Requirement Planning (MRP)

Push vs Pull systems
• Push system: initiates production in anticipation of future demand
◦ In an MRP system, lot sizes are determined based on forecasts of future demands and
possibly on cost considerations.
• Pull system: initiates production as a reaction to present demand
◦ In a JIT system, lot sizes are reduced to their minimum to eliminate waste and unnecessary
buildups of inventory.

MRP basics
• A production plan is a complete speci cation of the amounts of each item or nal product
and subassembly produced, the exact timing of the production lot sizes, and the nal
schedule of completion.
• Speci cation of the exact amounts and timing of production for each of the end items
• The production plan may be broken into several component parts:
◦ The master production schedule (MPS)
◦ The materials requirements planning (MRP)
◦ The detailed job shop schedule
• Phases:
◦ 1. Master production schedule (MPS): Specifies quantities of end items needed
◦ 2. Materials requirements plannning (MRP) system: Breaks down MPS and generates
planned order releases
◦ 3. Detailed job shop schedule: Develop detailed production schedile for the components or
subassemblies of each one of the end items
• From phase 1 to 3
◦ Bill-Of-Materials (BOM): How the products are broken down into components (subassemblies)
• Data sources for MPS:
◦ Firm customer orders
◦ Forecasts of future demand by item
◦ Safety Stock requirements
◦ Seasonal plans
◦ Internal orders from other parts of the organization

• What is MRP?
◦ A planning system that converts the Master Production Schedule (MPS) for nal products
into a schedule for assemblies and parts.
◦ Detailed schedule of production for each of the components that comprise an end item
◦ MRP is a tool to deal with these problems. It provides answers to several questions:
‣ What items are required?
‣ How many are required?
‣ When are they required?

Bill of Materials (BOM)

• It is a graphical representation of the relationship between the various levels of the productive
system. It incorporates all of the information necessary to implement the explosion calculus
• Weeks to produce end item = ∑ mayor L de cada level
• Calculate Net predicted demand (MPS) with demand and scheduled receipts:
◦ MPS = Demand per week - schedule receipts per week

MRP calculations level 1 (Lot for lot)

1. Gross requirements = MPS * cant. of component used of level 1
2. Net requirements = Gross requirements - on-hand inventory or scheduled receipts
3. Time-phased net requirements = Mover cant. de weeks de Leadtime del component level 1
atrás
4. Planned order release by period = Lot-for-lot (sacar lo que pides) (copy paste de lo de arriba)

MRP calculations level 2 (Lot for lot)

1. Gross requirements = MPS * cant. of component used of level 1
2. Net requirements = Gross requirements - on-hand inventory or scheduled receipts
A. Si on-hand inventory de semana inicio leadtime > gross requirements de semana 1 MPS =
Net requirements de semana 1 MPS = 0
3. Time-phased net requirements = production schedule = Mover cant. de weeks de Leadtime
del component level 1 atrás
4. Planned order release by period = Lot-for-lot (sacar lo que pides) (copy paste de lo de arriba)

Alternative lot-sizing schemes

EOQ lot sizing: planned deliveries y ending inventory

• Data needed: average demand rate, holding cost rate, setup cost
◦ 1. Calcular average demand = promedio de net requirements
◦ 2. Calcular Q = √(2Kd) / h = redondear para arriba
◦ 3. R = Leadtime * Demand = L * d = order when less
◦ 4. Ending inventory = Beggining inventory + Planned deliveries - Net requirements
◦ Gross requirements = MPS * cant. of component used of level 1
◦ **Planned deliveries y ending inventories se ponen en misma semana que gross
requirements
◦ Net requirements = Gross requirements - on-hand inventory or scheduled receipts
‣ Si on-hand inventory de semana inicio leadtime > gross requirements de semana 1 MPS
= Net requirements de semana 1 MPS = 0
◦ Time-phased net requirements = production schedule = Mover cant. de weeks de Leadtime
del component level 1 atrás
◦ Planned order release (EOQ) = Q en semana de inicio por Leadtime, volver a poner Q cada
vez que ∑ Time-Pased Net Requirements > Q
Comparison of Costs (EOQ vs Lot for lot)
• Setup cost = K * cant. de veces ordenadas
• holding cost = h per time unit * ∑ ending inventory
◦ En lot for lot = 0 (porque ordenas lo que sacas, no guardas)

Silver meal: average cost per period, number of periods the current order is to span and stopping
the computation when this function increases
◦ Fórmulas C(T)
◦ C(T) = Average holding and setup cost per period
◦ r# = Requirements
◦ C(1) = K
‣ If we order enough in period 1 to satisfy the demand in periods 1 and 2, then we must
hold r2 for one period
◦ C(2) = (K + h*r2) / 2
◦ C(3) = (K + 1*h*r2 + 2*h*r3) / 3
◦ C(j) = (K + 1*h*r2 + 2*h*r3 + ... + (j-1)*h*rj) / j
◦ Once C(j) > C(j-1), we stop and set y1=r1 + r2 + … + rj-1 and begin the process again,
starting at period j
◦ Se reinicia cuando C(j) > C(j-1) y volvemos a calcular desde fórmulas de C(1), ...
‣ y1 = the size of the order (lot size)
‣ Set first order lot as y1 = r1 + r2 = #
‣ Donde haces stop, el # de week donde empiezas después es # en y ... y# = r + r
‣ En el último periodo no volvemos a empezar, se deja y#(último periodo) = r (último
periodo) = #
 ◦ Respuesta: y = (#, 0, ... )
‣ # donde empezamos con C(1) otra vez
‣ 0 si el periodo es parte de una y#

Comparison of Costs (EOQ vs Lot for lot vs Silver Meal)

• Setup cost = K * cant. de veces ordenadas
• holding cost = h per time unit * ∑ ending inventory
◦ En lot for lot = 0 (porque ordenas lo que sacas, no guardas)

Lot sizing with capacity constraints: Lot shifting: Move demand from periods when demand
exceeds capacity to earlier periods when there is excess capacity
◦ r# = Requirements
◦ c# = Production capacities
◦ y# = Optimal production quantities
◦ 1. Solvable? (at least 1 solution exists)
‣ Anotar vectores r y c y calcular ∑
‣ Si ∑c ≥ ∑r = Solvable
◦ 2. Comparar por semanas c ≥ r
‣ Si c < r = sumar la diferencia a r de la semana anterior
◦ Respuesta: r' = vector c y cambiar por # de r en semanas que se sumó
a)

b)
c)
week 4 5 6 7 8 9 10 11 12 13 14 15 10 17
GrossRequirements from the mes 42 42 32 12 26 112 45 14 76 38
Net requirements 42 42 32 12 26 112 45 14 76 38
Timephased Netrequirements 42 42 32 12 26 112 45 14 76 38
PlannedOrderReleasenot for tot 42 42 32 12 26 112 45 14 76 38
week 2 3 4 5 6 7 8 9 10 11 12 13 14 15 10 17
vossRequirements from the mes 126126 96 36 78330135 42 228 114
cheduledreceipts
n handinventory
et requirements 126126 96 36 78330135 42 228 114
me phased Netrequirements 126126 96 36 78 330135 42 228 114
tannedOrderReleasenot for tot 126126 96 36 78 330135 42 228 114

se brincó la semana 33

component c

eek 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38
ossRequirements 165180300220 200240250290300
on the MPS

cheduledreceipts

n handinventory
et requirements 165180300220 200240250290300
me phased Net 165180300220 200240250290300
equirements
annedOrderRelease
of for tot 165180300220 200240250290300
component F H I

GIF 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38
GrossRequirements 330360600440400480500580600
from the MPS
scheduledreceipts

On handinventory
Net requirements 330360600440400480500580600

Timephased Net 330360600440400480500580600

Requirements

PlannedOrderRelease
not for tot 330360600440400480500580600

component F
week 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38
GrossRequirements 165180300220 200240250290300
from the MPS
scheduledreceipts

On handinventory
Net requirements 165180300220 200240250290300
ime phased Net 165180300220 200240250290300
Requirements

PlannedOrderRelease
lot for tot 165180300220 200240250290300

component E

week 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38
GrossRequirements 165180300220 200240250290300
from the MPS
scheduledreceipts

On handinventory
Net requirements 165180300220 200240250290300
ime phased Net 165180300220 200240250290300
Requirements

PlannedOrderRelease
lot for tot 165180300220 200240250290300
component

week 20 2 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38
Grossrequirements 330360600440400480500580600
from the MPS
scheduledreceipts

On handinventory
Net requirements 330360600440400480500580600
ime phased Net 330360600440400480500580600
Requirements

PlannedOrderRelease
tot for tot 330360600440400480500580600

component

week 2o 2 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38
GrossRequirements 495540900660600720750870900
from the MPS
scheduledreceipts

On handinventory
Net requirements 495540900660600720750870900
ime phased Net 495540900660600720750870900
Requirements

plannedOrderRelease
495540900660600720750870900
lot for tot

component

week 20 2 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38
GrossRequirements 165180300220200240250290300
from the MPS
scheduledreceipts

On handinventory
Net requirements 165180300220200240250290300
ime phased Net 165180300220200240250290300
Requirements

PlannedOrderRelease
165180300220200240250290300
lot for tot
 arribaviene

iii ii si iii ii ii ii aiii

CCD a 80

82
cal
Kthfrl 1001121422o

00 214271321 897.33
g ktilhltrajtalntele 1192114237

set 1st orderas

cis cal
I I 2 732 116

Ahora
ra no
142,42 3 ii 26 sí s 5E
a 1 K 80
21261
cal
kthztrl.ro 60

jt292
a.no 66
a3 ktilhllraftalhllh eet

I III
cal CRI set2ndorder as 44
eso 2

Ahora
384
no ra ra ra mo
42,4232 12 26 11 45 14 7
CHI K 80
54
cm
KIKI 180724244L

e.IE 7 137.33
a3 ktilhllraftalhllh
casi CRI set 3rd order as 47
13s
Ahora
4 s ra ra ra ra mo
142,4232 12,26 112 45 14 7 zaf
Asperiod 10 is the finalperiodin thehorizon we don't start the process again
i yo no 38

conclusion

y 116 0,0 150 0,0 135 0,0 381

42 42 32 12 Fink
100º
45 14 70 38 428
C 1100100,100 100 100 1,000
129190190,190
u r r r
100
V 1100 100 100,100,38 100 100 100 100

wenas 2 3
Grossrequirements 10 500 500 0
03 300
scheduledreceipts 96

On handinventory 186 60 30
o 66 36 78 33613542228
114
Net requirements o
Timephased Net 66 36 78 33613542228114
Requirements

PlannedOrderRelease 228114
66 36 78 336135 42
lot for tot

42 42 32 12 26 112 45 14 70 38 428
1.000
Fajajajajajajajajaja
1200200,200200,200,200 200 200
200,2007
 was 23
Grossrequirements

scheduledreceipts 96

On handinventory 186 60 30
o 66 36 78 336135 228
42 114
Net requirements o
Timephased Net 66 36 78 33613542228114
Requirements

PlannedOrderRelease 228114
66 36 78 33613542
Iot for tot

Links:
• Ejercicio excel: https://docs.google.com/spreadsheets/d/1-H2jC5kJjgw6oSB6Vz17cd-
D079VQcfvU8TH6z2w_BQ/edit?usp=sharing
• Handout: https://drive.google.com/file/d/1CWwJzcS9xdKkqaz4RN8kWdYIC32iJ8oi/view?
usp=drive_link