SALES OPERATIONS PLANNING

A definition of Supply Chain Management:

“Supply Chain Management is the process of planning (activities), implementing, and controlling (check the
order) the efficient, cost-effective flow and storage of raw materials, work-in-process, finished goods and
related information from point of origin to point of consumption for the purpose of conforming to customer
requirements” The Council of Logistics Management

Every company has its own supply chain.

Ex Manufacturing company  how many plants? How do you distribute the resources? Which is the market of
the finish product? B2B or B2C?

For example in Poland we need to go through a wholesaler to sell the product  contract with wholesaler.

The goal of Supply Chain is to satisfy customer requests, maximize customer’s service level

 The Right goods or services

 In the Right quantity
 To the Right place
 At the Right time
 And in the Right condition (quality)
 To the Right customer
 At the Right cost
Supply Chain & Operations processes

Pandemic impact  bad impact on demand forecasting

 Customer Relationship Management - provides the structure for how relationships with customers are
developed & maintained, including the product and service agreements between the firm & its
customers.
 Customer Service Management - provides the firm’s face to the customer and provides a single source
of customer information (Customer Service).
 Demand Planning – provides the structure for balancing the customers’ requirements with supply
chain capabilities.
 Order Fulfillment & Distribution Planning – includes all activities necessary to define customer
requirements and fill customer orders.
 Manufacturing Flow Management & Inventory Planning - includes all activities necessary to move
products through the plants & to obtain & manage manufacturing flexibility in the supply chain.
 Supplier Relationship Management - provides the structure for how relationships with suppliers are
developed & maintained.
 Product Development and Commercialization – provides the structure for developing and bringing to
market new products jointly with customers and suppliers.
 Returns Management – includes all activities related to returns, reverse logistics.
How to obtain a syncronized Supply Chain with S&OP?

Customer Service = Right Product in right Place at Right Time for the right cost
Customer Service
How to obtain a syncronized Supply Chain with S&OP?
1. Responsive Supply Chain (Leverage on production flexibility)
How many production batches? How many call-off to
suppliers/partners?
a. Supply Chain Synchronization
b. Agile vs Lean / Responsive vs Efficient
c. SC Risk Management & Resilience
d. Off-shoring vs Near-shoring
2. Accurate forecasts (improve forecast accuracy)
Which forecasting models? Which seasonality profiles?
a. SC visibility (demand planning vs sales forecast)
b. Matching demand patterns with forecast methods
c. New product introduction
d. Levereging Demand Sensing & Big data
3. Inventory (solve inaccurate forecast and inflexible production activities with inventory)
Which target inventory? How to size safe stock?
a. Inventory allocation (in the network)
b. Optimal safety stock 9holding cost vs availability)
c. Risk pooling and information sharing (VMI, CPFR, Beer Game)
d. Distribution Requirement Planning (DRP)

Wortmann classification of production / distribution systems

Customer Order Decoupling Point

 In the production-logic system there is a decoupling point that separates the part of the system
managed on demand from the part that has to be managed on forecast

The key components of the SCM / S&OP macro-process

Hierarchical levels of S&OP process

Supply Chain Planning Matrix: the software vendors’ view

S&OP / Supply Chain Planning

Supply Chain Planning workflow & 5 typologies of PLANS

Output of the planning process

Gaining competitive advantage through the supply chain

SCOR model for SC process mapping

Supply Chain integration

S&OP as a balancing act

Sales & Operation Planning: the junction box

Sales & Operation Planning: overview

Sales & Operation Planning: the process

Step 1: review and update the new product

development/introduction (NPD/NPI) programs, for
dissemination to sales and operations.

Step 2: create a new sales/demand plan, which is more than

just a forecast that includes input from all stakeholders in
demand planning.

Step 3: create the operations plan that will meet the demand
plan, ontime& in full, within known simple constraints. Match
the operations plan against major and critical resources, so to
sure the plans are realistic.

Step 4: integrate the numbers with the rest of the business.

What is the impact upon the financial numbers, people, IT
department, and more? Did we sell what we said we would
sell? Did we make what we said we would make?

Step 5: review the progress on actions from last time. Family-

by-family review; approve the recommendations; come up
with new proposals; make decisions; steer the business in the
right direction. Chaired by the top executives
Sales & Operation Planning: the process

Traditional vs. S&OP-driven planning

Volume vs. mix

How to model capacity constraints?

The S&OP planning horizon and granularity

THE S&OP CYCLE

THE S&OP DISTINCTIVE FEATURES

 Involves general management, marketing, sales, operations, finance, promotions and product development
 Includes Marketing Plan, Sales Plan, Promotion Plan, Production and Purchase Plan, Inventory Plan, Order
Backlog
 S&OP bridges business strategy with business planning
 S&OP’s output drives Departmental Plans
 The Planning Horizon should be sufficient to plan Labor, Equipment, Facilities, Material, and Financials required
to accomplish the Operations Plans
  Performed at aggregate level (Product Group/family)

THE S&OP CRITICAL SUCCESS FACTORS

Ongoing, Routine S&OP Meetings
Routine meetings on periodic basis (from quarterly to monthly). At least 3 meetings:
 Establish unconstrained demand plan & forecast
 Rough-cut supply plan & constrained demand plan
 fine tuning and alignment demand & supply plans

Structured Meeting Agenda

Structured Meeting Agenda Fixed Agenda with a pre-specified amount of time (e.g. 2 – 4 hours
meeting), in order to:
 Review of how well previous plans were met and perform a root cause analysis of plans
variance
 Discussion to align of demand side plans (Mktg, Sales) with supply side
 Closure of the meetings: single number planning environment

Pre-work to support Meeting inputs

 Baseline forecast, budget, orders, plant capacities, on-hand inventories.
 Data should be aggregated and translated for management
 Demand plan should include known impacts on future demand
 Supply plan should include known capacity problems, future shut-down, etc.
 Also external information should be brought into the process

Cross-functional participation
It is required an active participation during the meetings
 Demand side managers: Sales, Customer Service, Marketing
 Supply side managers: Manufacturing, Logistics, Procurement, Supply Chain
 Finance is involved to marry the operations plans with financial objectives Participants must be
empowered by the executives to make decisions

S&OP Management
The S&OP needs to be organised and run by a responsible organisation, in order to:
 Schedule meetings, Setting the agenda, Ensure pre and post meetings work
 S&OP manager must not dominate, rather drive to consensus

Collaborative process, leading to consensus

 As the output of the S&OP is a consensus-based plan, a collaborative process is needed
 Every stakeholder must be able to quickly create, review and revise plans

Joint Demand & Supply Planning to ensure balance

 A common mistake: to presume that sales & marketing plans are given and inflexible
i. Problem 1: Sales & Marketing has no role to play at the meetings
ii. Problem 2: Missing the opportunity to exploit excess in supply capacity
 Therefore also Marketing and Sales plans are rough-cut plans to be revised during S&OP

Process measurement
Like any other process S&OP should be measured so it can be improved over time
  Operational KPIs: performance measures: demand forecast accuracy, variance to baseline and
budget, customer order backlogs, plant utilisation, etc.
 Financial KPIs: SC costs and financial index (gross margin, cash-to-cash cycle, …)

Support of Integrated Demand vs. Supply Tools

A common problem: S&OP is supported by a myriad of asynchronous spreadsheets
 Step 1: implement a demand forecasting tool, since the main input of the process is the
baseline forecast (that should be generated by statistical tools, to keep it unbiased and
unconstrained)
 Step 2: integrate supply-side tools with demand side ones
DEMAND PLANNING
INTRODUCTION

Definition
“Demand Planning” is composed by a group of Supply Chain processes, management methodologies and
statistical quantitative models to enable the definition of the Demand Plan for manufacturing or distribution /
logistics companies.

Demand Planning involves all the business processes whose objective is

 The generation of sales forecasts for the products / customers groups

 The generation of sales budget, mixing forecasts, orders and marketing plans
 The definition of constrained demand plan, feasible and compliant with the limited capacity
resources of production, supply and distribution

Forecast aims at:

 estimating the probability of future events

 determining time & location of future events
 predicting importance of random variables in future
 identifying the regularity on historical data
 pointing out the connection among quantities measured in the past

Demand Plans represents the main input for the other Supply Chain Planning processes concerning:

– production
– purchasing
– distribution

Demand forecasting & planning process

What? Aggregation levels

When forecast
– Long term (>2/3 years): STRATEGIC DECISIONS (planning about BU, product line, market)
o Total sales, production capacity, distribution model, new products launch, …
– Mid term (1/2 years): TACTICAL DECISIONS (annual budget; aggregate forecast)
o Total sales by product line, prices by product line, economic conditions, …
– Short term (0 / 6 months): OPERATING DECISIONS (weekly or monthly disaggregate forecast)
o Sales by item, geographic area, customer, prices, volume ...

It presumes that demand for a

product follows a S-shaped curve
growing slowly in the early stages,
achieving rapid and sustained growth
in the middle stages, and slowing
again in the mature stage.

How? Forecasting models

 Qualitative (judgmental) forecasting techniques:
o Sales force
o Expert evaluation / delphi method
o Possibility / analogy
o Market research
  Causal methods :
o Regression (linear, square, multiple,...)
o Econometric/ input-output
 Time-series forecasting techniques :
o Moving average (simple, weighted,..)
o Exponential smoothing (winters...)
o Arima (box jenkins)

DEMAND PLANNING AND S&OP PROCESSES

Sales & operations planning: definition (recap)

Sales & Operations Planning includes Supply Chain processes related to the planning of Sales, Production and
Distribution activities across Supply Chain Networks multi-level

Sales & Operations Planning processes:

 Medium term processes

o Demand Planning, Material & Capacity Requirements Planning (MRP / CRP), Distribution
Requirements Planning (DRP)
 Short term processes:
o Operations Scheduling (job loading, allocation & sequencing), Transportation Scheduling
(vehicle loading &routing)

Sales & operations planning: objectives (recap)

The objective of the macro-process called S&OP (Sales & Operations Planning) is the generation of feasible
operation plans, using in an optimal way the available business resources (production, distribution and
inventory capacities) to satisfy the market demand for the company’s finished goods.

Analyzing:

1) Availability of resources along time periods of the planning horizon

2) Sales demand by product, country / customer, period the S&OP process aims to determine:
• Demand plans: trade promotions, sales budget, …
• Supply plans: production plan, purchasing plan, logistics transportation plan
Importance of demand planning processes
Critical factors related to products
• Number of SKUs
• Product mix (assortment planning)
• Product life cycle
• Marketing actions (promotion planning)
• Product shelf life (ex: seasonal collection)
• Critical components purchasing lead times

Critical factors related to markets

• Number of B2C / B2B customers
• Customer service level
• Competitors’ commercial actions
• Delivery lead times
• Sales channels behavior (retail, wholesale)

SELL-IN VS. SELL-OUT

 Sell-Out
o Final consumer demand (client), estimated by POS manager
o Final B2B customer demand (manufacturing company), which purchases components / WIP
from a manufacturer (supplier), for assemblying then components into more complex finished
goods
 Sell-In
o B2C: POS demand, estimated and filtered and modified by distributor (wholesaler modifies the
retail Sell-Out demand, adding safety stock and following suppliers’ price discounts)
o B2B: Finished products demand, estimated by the manufacturer (supplier)
 Differences between Sell-In and Sell-Out
o Manufacturer → Distributor: additional stocks (pricing), safety stock
o Distributor → Point of Sales: local promotions (ex: discounted price)
o Sell-Out: forecast made by distributors, store managers, …
o Sell-In: demand plan generated by distributor (client) to the manufacturer
o The manufacturer estimates Sell-In demand (Sell-In forecast)
 o Safety Stock: 10% of Sell-Out demand

Importance of demand forecasting

• Under-forecasting
– Customer service level reduction
 Potential periods measuring stockouts
– Safety stock level increased
– Continuous review of demand & supply plans
– Reduced sales and market quota
• Over-forecasting
– Inventory holding costs increased
– End of life / shelf life (physical or technology obsolescence)
– Capacity allocation

Under-forecasting
• Actual sales have been underestimed by 20%, in all the past months
• Actual service level: 80% (240 pieces lost)
Over-forecasting
• The real customer demand (actual sales) has been overestimated by 50%, in all the historical months
• Actual service level: 100% (0 lost sales units)
• Overstock final level: 600 pieces; 50 pieces on average
TYPOLOGIES OF DEMAND PLANS

Demand plans

Sales Forecast
 Quantitative forecast of the future demand
 Typically prepared using quantitative Time Series methods

Sales Plan
 Define the commercial plan for the sales; it involves different sub-plans:
o Statistical forecasting (Sales Forecast)
o Marketing actions (Trade Promotion Management)
o Sales Plan prepared by sales rep or channel managers

Demand Plan
 It is the Sales Budget «feasible», that is: constrained to the distribution and production constraints
(Supply Planning)
o Production, distribution, storage, supply constraints

Sales Forecast vs Sales Plan

 Slow movers / obsolete vs. best seller / class A products
 Old vs. new innovative items
 Forecast vs. Order Netting, salesforce constraints
 Promotion actions (increase of the forecast baseline)
 Top Management & Finance inputs (revenue target, profitability)

Example 1
Given a product – market couple, in a time bucket t

 Sales Forecast SF = 10 pieces

 Delta promotion DP = +15% of Sales Forecast
 Portfolio orders OA = 7 pieces

Sales plan = SB

SB = SF + sup (DP) + OA  10 + sup (1.5) +7 = 19 pieces

SB = max (SF; OA) + sup (DP) = 10 +2 = 12 pieces

Example 2

 Maximum value, by period, between orders OA and forecast SF

 Delta Promo expressed into sales quantities (not percentages)
 The user (sales manager) can correct the forecast plan (mix of statistical forecast, orders, promotions)
 SKU1 / Jan05 → estimated effect of cannibalization on SKU1 of promo action on SKU2 / Jan05
 Management intervention on SKU2 / Mar05: sales reduction due to obsolescence (end of lifecycle)
 Management intervention on SKU1 / Mar05: sales increase due to half-year budget constraint (new
promo action to be defined for reaching the expected budget)
 KPI: punctual difference between forecast SF / budget SB

So…

 The calculation of the Sales Forecast has to be done before, from a Demand Planning workflow point of
view, the Sales Budget definition
 Sales Forecast is the first input to the Sales Budget, it defines its volume size and its time profile
 Adding to the Sales Forecast the delta promotional and netting the customer orders currently in the
Order Portfolio, we obtain the plan to submit to the Management approval, delivering as output the
Sales Budget
 Often and incorrectly, companies define Sales Budget in an independent way (and previously) from the
Sales Forecast, not leveraging the powerful information value added inside the statistical forecasting
plan

Sales budget vs. Demand plan

Demand Plan
  It is the output of the collaborative process of negotiation and balancing of the Sales Budget between
“Demand” and “Supply” functions.
 The Sales Budget is submitted to the feasibility analysis with the “Supply” functions (production,
distribution), comparing:
o Demand: sales budget (company profitability)
o Supply: production capacity constraints, inventory capacity, distribution / transportation
capacity, purchasing budget constraints
 The Demand Plan (constraint plan) is production and logistics feasible, satisfying all the “supply”
constraints; it represents the official final plan for internal diffusion to the company functions and to
the external Supply Chain partners

The production has to be scheduled (for shelf-life constraints) in the same month when demand occurs (Sales
Budget): no possibility to anticipate or delay the production quantities

Production
constraint: maximum
month capacity

This constraint is not

respected on Mar05
for the item SKU1
DEMAND PLANNING PROCESSES & WORKFLOW

Demand analytics

 Hierarchical analysis OLAP-based (On Line Analytical Processing) / historical sales analysis
 Sales analysis for performance evaluation
 Forecast Accuracy calculation
Sales forecasting
 Sales Cleaning: outliers removal from historical demand patterns
 Best Fit on parameters (Holt-Winters)
 Sales Forecast Generation

Marketing intelligence: products

Actions on products
 Define the marketing and development plan on products
 Define promotions on slow-movers low-sales products, discounting ales prices
 Define conjoint offering of couples of groups of articles, with the same packaging, sold during special
events or marketing events
 Update the price lists for finished goods
 Define marketing campaigns on specific product groups or lines, using proper multimedia advertising
channels
 Prepare commercial catalogs to be distributed on POS, accurately projecting the layout of catalog or
Web pages
 Define location of products on POS shelf (store layout planning)

Marketing intelligence: customers

Actions on markets
 Evaluate new opportunities to open new stores / POS at new geographic areas
 Evaluate sales perspectives on potential prospects
 Define new promo actions (contents, duration, presentation & communication, economic parameters
of the offer) on item/customer clusters
 Introduction of new item lines on specific target markets
 Project of commercial relationship of collaboration with single customers (wholesalers B2B),
integrating logistic plan (Vendor Managed Inventory, CPFR)
 Definition of the contact way with consumers in B2C (web sites, mailing list, multimedia advertising,
call centers for customer care activities)
Demand intelligence

Data mining
 Clustering (on groups of items or customers / stores)
 Sequence clustering
 Classification (ABC, promo, ...)
 Association rules (market basket analysis)

Role of demand planner into the organization

Typical tasks
 Forecasting Ownership (forecast champion)
 Input data (quantitative, qualitative) research
 Demand Analytics (KPI monitoring)
 Preparation and control of Demand Plans
 Human interface to company’s key users and external business experts
 Organize periodic S&OP / DP meetings
 Collaborative inter-functional negotiation
 Distribution of demand plans to the functions

Key competencies
 Math – statistics: knowledge of Sales Forecasting / Data Mining algorithms and big data analysis
methods
 Relational – communication
 Organization: managers during the meetings, identification of best functional resources (forecast group
definition)
 Business knowledge: expertise on markets where company operates

HISTORIAL SERIES: INTRODUCTION

Time series
A time series is a sequence of measurable data (orders, euro, kg, litre) points (D1 , D2 , D3 , … , Dt , …), obtained
typically at successive times, spaced at (often uniform) time intervals (days, weeks, months, quarters, years, …)
Time-series forecasting techniques
 CONTEXT: planning (stock management, distribution plan, production plan, etc.)
 DETAIL LEVEL : item, SKU, family
 FORECAST HORIZON : short-very short term (1-6 months)
 HISTORICAL DATA :
o weekly/monthly/two-month period sales
o at least 2 years history data are available (for seasonality)
o Continuous and predictable demand (variation coefficient : sigma * /DM)
 ASSUMPTION : future will be like the past

How? Multiple methodologies’ usage

Time series analysis

Objectives of the Time Series Analysis

 Outliers detection
 Historical series component analysis
o Basic typologies
 Continuous series
 Sporadic series
o Components
 Seasonality (default length: 1 year, 1 sales cycle)
 Trend (linear, nonlinear)
 Cyclical (long term trend)
  White noise (erratic unforecastable component)

Demand components

Before start, forecasting for the future it is necessary to analyze historical

data to identify  trend and seasonality
Types of time series (without seasonality)

Types of time series (with seasonality)

Demand components: seasonality

Seasonality causes
 Climatic
 Customs (traditional recurrences, holidays, ...)
 Cyclic sales promo (fashion weeks...)
Extra-seasonal occurrences
 Accounting & fiscal (budget, …)
 Effective working / trading days
 Calendar (ex. American : 4-4-5)
 Moving holidays (ex. Easter)

Forecast-ability: decision tree about pattern

Historical series forecast-ability

How easy is to forecast a historical series?
Which methods have to be used?

First check: new items vs. permanent items

Second check: continuous series vs. sporadic series
Third check: regular patterns on continuous series

 If a historical series contains at least 50% zeroes, it can be

considered having a sporadic behavior;
 It has to be performed, in addition, a control about the shape
of the series. Example: items sold or promoted into specific
periods of the year (sport events, racing competition, etc.) are
characterized by high sales «in season» and zero sales «out of
season» → they must be processed using Time Series
forecasting methods (removing «out of season» periods with
no sales)
Historical sales typologies
We distinguish at least 2 different typologies of demand patterns:

1. Items having medium – high demand in the time horizon

(continuous items) ex. CPG, mass market goods
 2. Items having low and intermittent demand (“lumpy”) in the time
horizon (sporadic items) ex. Spare parts, class C items

FORECAST ACCURACY
Forecast error in period t is
Historical series intervals: the difference between
 Training Set actual demand and
 Test Set (forecast calculated over the past) forecast for that period
 Forecast horizon (in the future)

KPIs
Time extension

 Punctual measures (calculated only on one time bucket)

 Global measures (calculated over a time interval)

Units of measures

 Absolute measures (in quantity or values)

 Relative measures (percentages)

Usage

 Parametric Best Fit (choice of best parameters in Holt-Winters)

 Model Best Fit (choice of best algorithm)
ME: Mean Error
 Indicate BIAS :
o ME< Average Forecast
o ME>0 : Average Demand > Average Forecast
 It allows positive and negative errors to partially offset each other

MAD: Mean Absolute Deviation

 it measures absolute (without regard to sign) error consistency
 opposite errors do not compensate each other
 it does not show errors’ correlation

MAPE: Mean Absolute Percentage Error

 It allows to compare different series on a % scale
 It penalizes more errors in periods with a little demand
 It doesn’t work with null values of demand

MAPE CALCULATION

SDE: Standard Deviation of Errors

 By squaring the individual error SDE penalizes more high absolute errors than small errors
 It refers to n values (n-1 represents the number of data independent from each other)
 Often used for estimating safety stock

WHICH ERROR METRIC TO USE?

For measuring forecast error, it is correct to use:

 MAPE: Mean Absolute Percentage Error

o for comparing performances among SKUs, countries, etc.
 o To immediately understand how far is your forecast from actual
 MPE: Mean Percentage Error
o for highlighting the percentage magnitude of BIAS error (underestimation or overestimation of
demand)
o This KPI has to be analyzed secondly, in case of high value of MAPE

MAPE and MPE are not the proper metrics when you want to measure forecast error in presence of sporadic
series (i.e.: series having majority of zero values)

MAPE CONTROL
Setup of the threshold values for MAPE (alerting)

 Based on vertical industry benchmarks

o Example: in the Food & Beverage market, a sales forecast over a couple SKUL / all (country) /
month having a MAPE not greater than 10% can be considered (almost) excellent (best in class)
 Based on continuous improvement
o To reduce progressively the threshold value of acceptance for the forecast (25%, 20%, 15%, …)
 Based on ABC Classification (Pareto on Revenue)
o MAPE < 15% → class A items
o MAPE between 15% and 25% → class B items
o MAPE between 25% and 35% → class C items

SALES CLEANING

For continuous items

General workflow
STEPS
1. PERIODS: Identify which periods have to be analysed and, eventually, cleaned
2. ALGORITHMS: Define the parameters for setting-up the chosen cleaning method
1. Moving averages → lenght of average
2. Single Exponential Smoothing → alpha parameter for smoothing outliers
3. Confidence Interval → number of standard deviation
3. PRE-CLEANING: remove seasonal & trend components to the historical series (still to be cleaned)
4. CLEANING:
1. Certify whether a value has to be cleaned or not
2. Clean the certified outlier
5. POST-CLEANING: rebuild seasonal & trend components to the historical baseline (already cleaned)

Confidence Interval
One-step method
Cleaning of all historical periods

Two-step method
 Cleaning of non promo periods
 o The historical series S1 to be cleaned in phase 1 is composed by only the non promo historical
periods (the promo periods are temporary removed from historical series)
o Phase 1 Sales Cleaning is run for removing from historical series S1 outliers related to pas
stockouts, outliers characterized by abnormal sales (big customer orders), statistical «white
noise» (all the outliers that are not linked to past promo events)
 Cleaning of promo periods
o In Phase 2, the historical cleaned S1 series is relinked to the series of non promo periods (still
not cleaned), rebuilding an historical series having only non promo periods already processed /
cleaned
o Phase 2 Sales Cleaning applies only over promo past periods, to remove the only effect of trade
promotions (which can be now better identified from a statistical point of view, after Phase 1
application)

So..

 Phase 1: from “Historical Demand” series (blue) statistical no promo outliers are removed, obtaining
“(1) No promo Cleaning ” (red) output series;
 Phase 2 : from “Historical Demand” (only promo periods), in union with “(1) No promo Cleaning” (red)
(promo period already cleaned), we obtain the final historical baseline series «(2) Promo Cleaning»
(green).
SALES FORECASTING: EXPONENTIAL SMOOTHING MODELS

Time series forecasting techniques

 Moving average models
 Exponential smoothing models (single, double, Holt-Winters)
 Time series decomposition models
 Sporadic series models
 New item introduction models

Historical series analysis

Seasonality component
In this case you have 12 seasonal indices, one for each seasonal cycle Sjan , Sfeb , Smar , …, Sdec one for each
month.
Plotting seasonal indices you obtain a seasonality graph.

Trend component
You have to evaluate the monthly time series below :
Time series analysis (trend)
With regression analysis it is possible to identify and evaluate the characteristic of trend

It is necessary to identify the theoretical function y=f(t) (straight line, parabola, …) which better fits historical
data series

Assuming a linear type of trend it is necessary to

determine the values of regression line’s coefficient.

IN THIS CASE IT RESULTS :

THE EQUATION OF REGRESSION LINE IS

Forecasting terms
 TIME BUCKET : t coincides with how
often the plan in updated
 ACTUAL DEMAND in period t : Dt
 FORECAST about period t+m made at
the end of the period t : Ft+m
 TIME HORIZON : is the time frame for
the plan → m

SIMPLE MOVING AVERAGE

Given a time series of ”n” values (dt , dt-1 , dt-2 ,…), at the end of the period t you can calculate the value of
moving average (“k” is the moving average time lag ) :

Ending to the last available period you obtain “n-k+1” values for the moving average
In the example: N=24, k=4 21 values for the moving average with time-lag 4

ALTERNATIVELY WITH k=6 MOVING AVERAGE YOU OBTAIN 19 VALUES

The irregularity of the time series can be filtered by the moving average

NUMERICALLY YOU OBTAIN:

Forecasting model
If time series is stationary and
non-seasonal, moving average
can be used to forecast the future
demand
SINGLE EXPONENTIAL SMOOTHING
BROWN’S MODEL
Given demand time series D1 , D2 , … , Dt , t+1 period forecast is:

Forecast is obtained by the weighted average of actual Dt and previous forecast Ft

Iteration process:

Brown’s model (simple smoothing) is suitable for data with no trend

or seasonal patters

Forecast Ft+1:
 Belongs to interval { Dt ; F t }
 Needs only two data f(D t , F t )
 Holds all historical data (Dt , Dt-1 , …, D1 ) weighted with decreasing values following a negative
exponential function

α value influences model’s reactivity

Example: α=0,5
To each element is assigned a different weight based on
seniority
  α high : reactive model (> weight to new data)
 α low : static model (> weight to past)

Example

Calculate sales forecast from the time series below:

Using brown’s model, with both alternatives: α=0,3 and α=0,5

Finitial = 140

Once the process starts, the model calculates forecast month by month:
Carrying on the calculations it’s possible to simulate the forecasts for each month in the past 2 years (24
months)

Proceeding in this way it is possible to generate forecast for both 2 years and for January 2004
Series decomposition
multiplicative model
 TREND: linear regression
y(t) = a + b·t
 SEASONALITY: seasonal
profile S(t)
 FORECAST: F(t) = (a +
b·t)·S(t)

Implementation: future forecast

Thanks to simulation results, it’s possible to project forecast in the future.

At the end of
simulation period, forecast is generated for the next 6 months (with the optimal model’s configuration)

Once you have new demand data you can update model’s variables in order to project forecast in the future

At the end of month 13 forecast is recalculated for the next 6 months (“frozen period” is only month “+1”)
ADVANCED FORECASTING METHODS
• CROSTON’S MODEL
• POISSON’S MODEL
• NEW ITEM PLANNING

Demand Planning for Slow Moving items

Sales patterns:
• Sales events are rare and completely casual, not correlated among them
• Sales orders are for unit volumes / few quantities; they occur rarely in the planning horizon
High-tech manufacturing context:
• Maintenance Repair and Overhaul (MRO)
• Spare Parts
Methods:
• Croston → forecasting algorithm
• Poisson → replenishment algorithm

Inventory Planning for Slow Movers items

Hypothesis on demand patterns

• Demand patterns are irregular, having low values, in a timely sporadic way, without causal
relationships
• If an item is sporadic in the past → it will remain sporadic also in the future planning horizon
(Forecasting hypothesis)
• Variability range for sporadic series in manufacturing retail is from 0 to 10 pieces per period
• Example of series: {0 0 0 1 0 0 0 0 1 1 0 0 0 0 2 0 0 1 1 0 1 0}
o no trend and no seasonality can be «identified»
CROSTON’S MODEL

Introduction
• Method developed in 1972
• Used for time series forecasting model for sporadic items
• It provides answer to the following questions:
o Which is the next positive demand volume?
o When next positive sales will occur?

Algorithm
The time series is composed by two indipendent parts, to be separately forecasted:

1) Historical series of positive volumes

2) Series of time interval between two consecutive positive values of sales

How Croston’s algorithm works

• Given the series of positive sales, we calculate the volume Q of the future demand (a unique value Q
for each sku – warehouse couple), applying forecasting algorithms such as simple moving average (MA)
or single exponential smoothing (SES) over the sequence of positive historical values only

• Given the series of time interval between 2 consecutive sales (integer number of time periods), we
calculate the forecast of the next «average» interval lenght T applying forecasting algorithms such as
simple moving average (MA) or single exponential smoothing (SES) over the sequence of delta intervals
collected into the historical series

Example: Historical demand: {10 0 0 0 4 0 0 2 0 0 0 0 0 3 0 2 0 0 1}

Positive volume series: {10 4 2 3 2 1}; average: 22/6 = 3,7 = Q

Delta interval series: {4, 3, 6, 2, 3}; average: 18/5 = 3,6 = T

We put the future fixed forecasting volume Q

• Rounded (→ if sales are discrete volumes)

• Not rounded (→ otherwise)

in a regular «semi-periodic» way (following the length of the average interval T), both over historical periods
and over future periods of the planning horizon:

• Starting from the oldest positive historical value of demand (over which we put – by default – the first
future demand volume Q);
• Calculating – at every step – where positioning the next Q value in the time axis, using a simple
algorithm «rounded integer»

Example
• Historical monthly demand: {10 0 0 0 4 0 0 2 0 0 0 0 0 3 0 2 0 0 1}
• Delta interval series: {4, 3, 6, 2, 3}; average : 18/5 = T = 3,6 months
• Demand forecast: {0 0 Q 0 0 0 Q 0 0 0 0 Q …}; in the example: Q = 22/6 = 3,7 units (rounded at 4 units)
 • Where to place Q volumes:
o Step 0 : Q is placed on historical volume 10 (the oldest historical value)
o Step 1 : int(3,6) = 4 months (→ Q is placed on historical volume 4)
o Step 2 : int(3,6–0,4) = int(3,2) = 3 (→ placed on historical volume 2);
o Step 3 : int(3,6+0,2) = int(3,8)= 4 (→ placed on ninth 0)
o Step 4 : int(3,6-0,2) = int(3,4) = 3 (→ placed on eleventh 0)
o Step 5 : int(3,6+0,4) = int(4) = 4 (→ placed on historical volume 1)
o We proceed this way also for all future months
• int(…) = nearest integer

Numeric examples
• Historical volumes: October 2004 to September 2005
• Future months: October 2005 to September 2006

Sporadic series with high volumes

• Historical demand: {10 0 0 0 100 0 0 2 0 0 0 0 0 250 0 50}
• Delivery lot size (or production lot): 50 units
• The manufacturing company delivers:
o to some customers: multiple lots
o to other customers: any volumes (also small / unit lots)

Solution modeling:
• Distinguish sales between «big» and «small / medium» customers
• «Small / medium» customers are managed following the standard Croston’s / Poisson’s algorithm,
without pre-processing the numerical volumes of historical demand
• For «big» customers, after nullifying small quantities eventually included into the historical series (i.e.
whose value is strongly less than multiple lot size), we divide the total «big» quantities by the multiple
delivery lot size, obtaining a new historical series that is sporadic with small positive (integer) values →
in the example: {0 0 0 0 2 0 0 0 0 0 0 0 0 5 0 1}
POISSON’S MODEL

Inventory Planning for Slow Movers items: Poisson’s model

For solving the manufacturing retail planning problem for spare parts we use the statistical distribution of
Poisson
Poisson’s distribution: numerical histograms
Poisson’s distribution
o Lead Time: is the delivery time to the «target» location (Wartsila central warehouse), that is: the time
interval between two consecutive deliveries
o D(LT) is calculated as average of the historical demand (after aggregating historical sales for each
historical LT), including into the average calculation also periods having zero sales

Poisson’s distribution as sales forecasting model

 Using it as a Sales Forecasting model, the Poisson’s distribution provides only a probabilistic range
about future sales quantity (ex: the probability to sell 3 pieces in the next LT = 2 weeks is 87%)
 This info is only useful for strategic planning scenario analysis (long term planning models)

Poisson’s distribution for manufacturing retail planning

Poisson methodology steps

Step 1

 Assign a target service level SL for each SKU – customer or SKU – warehouse combination, via the
Service Level matrix grid (ex: 99,5%, 98%, 95%, etc.)
 Calculate the INVMIN minimum stock level to hold in warehouse to satisfy the required service level
SL, supposing that the combination follows (in the past sales but also in the future horizon) a
distribution very close to the Poisson’s one
 Which is the minimum stock level of pieces (INVMIN) to hold in stock to have a «success probability»
(→ not to have stockout at all during the lead time LT) at least equal to the target service level?
 The INVMIN calculation is done by cumulating each success probability / event. Example: if I hold
INVMIN = 2 pieces on hand, there’s no stockout is the actual demand is either 0, or 1, or 2.

Step 2

If we suppose to have, at the beginning of each delivery lead time, a future demand equal to INVMIN level,

 for each lead time interval in which the planning horizon can be divided, starting from «today» (time
now) (ex: 6 different lead times of 3 weeks, in a planning horizon lasting 18 weeks)

if we replenish, at the beginning of each consecutive lead time, the stock needed to satisfy the INVMIN
demand,

 we will guarantee the target service level SL, regardless the specific time period (day, week) internal to
the lead time during which the sporadic demand will fall.

Step 3

From the classification of Replenishment methods point of view (warehouse replenishment), the Poisson’s
algorithm can be considered as min-max replenishment methodology:
  if the current stock is lower than INVMIN,
 we need to reorder a variabile quantity to cover a target stock equal to INVMAX

where: INVMIN = INVMAX = minimum quantity output of the Poisson’s algorithm

• Minimum stock level INVmin → if current stock is lower than INVMIN we launch a new purchase
order;
• Maximum stock level INVmax → stock level to reach when preparing a new purchase order

Step 4

(INV is the on hand stock, at the beginning of each lead-time)

Example 1

• if INV = 2 and INVMIN = 3 → replenish 1 piece

• if INV = 0 and INVMIN = 3 → replenish 3 pieces
• if INV = 5 and INVMIN = 3 → not to replenish (and recalculate the stock projection at the end of the
next lead time)

Example 2

• if INV = 0 and INVMIN = 2 → replenish 2 pieces

• if INV = 1 and INVMIN = 2 → replenish 1 piece
• if INV = 2,3,4,… and INVMIN = 2 → not to replenish

Safety stock calculation in Poisson’s algorithm

The safety stock calculation is «implicit» inside the Poisson’s method for calculating the INVMIN stock level.
The safety stock is the difference between the two values:

• INVMIN Poisson’s level (both target stock level and reorder point)
• Average historical demand during the delivery lead time

Example

• if average demand D(LT) = 2,4 pieces and INVMIN = 3 pieces

• then the safety stock is equal to: SS = INVMIN – D(LT) = (3 – 2,4) = 0,6 pieces

Poisson’s distribution: exercise

Based on historical sales data of the last years, we calculate that the average demand for the item “Alpha
0101R65” into the 0101 warehouse has been equal to 1,2 pieces/week. Supposing that the delivery lead time
is equal to 2 weeks, and the required SL is 75%, calculate:

• The probability that the item “Alpha 0101R65” will be sold into the next 2 weeks
• The probability not to have stock-out supposing to have on stock (on hand) 3 pieces of the item
 • The expected number of stock out pieces in 1 year (supposing that the order lot size Q is equal to 3
pieces)

Supposing an annual inventory holding cost equal to 25% and a unit stock-out cost equal to the 40% of the
product selling price (250 €), calculate, the inventory management global costs for the replenishment
configuration

Step 1: we need to reconduct the average weekly demand value to the average demand during the lead time
(lasting 2 weeks):

Step 2: we calculate the Poisson’s distribution values for

different future possible demand values k = 0, 1, 2, 3…

We obtain in output the following Poisson’s probability distribution

for the future demand values k = 0, 1, 2, …

We can also evaluate the stock coverage demand

level: if we keep on hand 3 pieces of leather goods
item “Alpha 0101R65”, the probability not to have
stock-out during the next lead time (2 weeks in the
planning horizon) is equal to 77,8%:

NEW ITEM PLANNING

Problem definition
• To forecast the future sales of new items, not having for them – by definition – historical sales available
• To forecast rapid changes in demand patterns

Application fields
• Pre-season forecasting in the sector fashion / retail (sales of the first launch of a brand new product)
• New models introduction in high-tech (consumer electronics, information technology), becoming
rapidly obsolete
• In general: all the items having a limited lifecycle (at maximum, some months of lifecycle, only one
sales “season”)
• Changes in market behavior (crisis, consumer spending review, fashion)

Product lifecycle curves / patterns

One-seasonal product
Launch; increasing sales, maturity; decline
Multi-seasonal product
Launch; increasing sales, stable sale

METHODOLOGIES
Lifecycle curves (percentage patterns)
Pattern matching algorithm

• Initially we build a pattern library (percentage or absolute series of numerical values with time flag
identifier: sales percentage of the first week, of the second week, etc.) based on historical values of
new item introduction occurred in the past
Example: curve XYZ – product 123 – market ijk – percentage values for the first 7 weeks of the sales
cycle: {15%, 15%, 20%, 25%, 15%, 5%, 5%}.
• The initial sales pattern for the new segment or couple product – market is chosen, manually or using
automatic criteria (ex: highest similarity, same family, same channel,), inside the pattern library
previously built inside the Sales DataMart.

Example

In-season budget (defined by

Marketing / Sales): 15000 pieces

Seasonal campaign duration for the

item NEW into the store STORE_A: four
months (March to June)

Analogy forecasting (product links)

• The forecast for the new couple product – client is calculated using classic Time Series algorithms,
playing as the new couple had an historical pattern of sales.
• The historical sales patterns of the new couple is “created” by linking the historical series of an old
product/client to the new one.
• The underlying hypothesis of this method is that the customers will perceive the new product sold in a
given market as “substitutional” (totally or partially) of an obsolete one, progressively removed from
the market offer of the company

• Single or multiple link

o Link 1:1 → 1 new product / 1 obsolete product
o Link 1:n → 1 new product / n obsolete products
• Linearity of the link
o Linear: the historical sales of the old product are copied “as is” over the new item phantom
«historical sales»
 o Not linear: to the historical sales of the old product, we apply multiplicative coefficients for
increasing (new product is better than old one, or better promoted) or decreasing historical
sales (the new item is only partially substitutional)

Example

NEW = 100% SKU1 + 50% SKU2

End of lifecycle SKU1: Feb-05

End of lifecycle SKU2: Apr-05

Start NEW: Mar-05

TIME SERIES FORECASTING MODELS

HOLT – WINTERS ALGORITHM

HOLT’S MODEL
• At the end of period tis possible to calculate forecast for the generic period in the future t+m
• Forecast is obtained starting from smoothed level lt , corrected by smoothed trend tt and respective
time horizon m
• m : forecast horizon

at the end of period t, disposing of new actual datum, it’s possible to update
level and trend smoothed values:
Level and trend updating follows exponential smoothing is principle

Summary:

• Holt’s model is ) is suitable for data with trend but without seasonal patters (or to unseasonalized data)
• Forecast Ft+m
o needs 3 data f(D t , L t-1 , T t-1 )
o Holds all historical data (Dt , Dt-1 , …, D1 ) weighted with decreasing values
•  and  values influence model’s reactivity
• To m increasing corresponds a reduction of forecast accuracy
• It’s possible to decrease trend component projected in future by F [0 ;1] parameter

Example : apply Holt’s model to the time series in the chart below:

1. Given initial values L1 =90 and T1 =5 it’s possible to update by two-month period the value of Lt and Tt
(adopt initially =0,3 e  =0,5)
 2. At the end it’s possible to calculate forecast month by month (m=1 for each two-month period and
then (from the 6th two month period) it’s possible to calculate forecast for the future (m=1, 2, 3, 4, ...)

HOLT-WINTERS MODEL
At the end of period t it’s possible to calculate forecast for generic period in future t+m :

Forecast is obtained starting from smoothed level lt , corrected by smoothed trend tt and respective time
horizon m, adjusted with smoothed seasonal index st+m
AT THE END OF PERIOD t IT’S POSSIBLE TO UPDATE SMOOTHED VALUES OF LEVEL, TREND AND SEASONALITY:

Summary:

• Winters model is appliable directly to a time series with trend and seasonality
• Forecast Ft+m:
o Needs 4 data f(Dt , L t-1 , T t-1 , S t+m-L )
o Holds all historical data (Dt , Dt-1 , …, D1 ) weighted with decreasing values
• α, ß, γ, values influence model’s reactivity
• By the increasing of m the accuracy decreases

IMPLEMENTATION: FUTURE FORECAST

Thanks to simulation results, it’s possible to project forecast in the future.
At the end of simulation period, forecast is generated for the next 6 months (with the optimal model’s
configuration)

Once you have new demand data you can update model’s variables in order to project forecast in the future

At the end of month 13 forecast is recalculated for the next 6 months (“frozen period” is only month “+1”)
MULTIPLICATIVE SERIES DECOMPOSITION: THE ALGORITHM

Decompose an historical series into trend / seasonality / cyclicity components (multiplicative model):

1. Calculate the conjoint component of trend T and cyclicity C by applying a centered moving average (CMA)
algorithm, given L step equal to seasonality lenght:

2. Calculate the seasonal S component using multiplicative seasonality coefficients:

3. Calculate the mean seasonality coefficients: removal from historical series of
random fluctuations component et to isolate seasonality, averaging St  et values over homologue seasonal
periods → output: SMt
4. Deseasonalize the historical series values by dividing each demand value by the corresponding seasonal
coefficient SMt:

5. . Calculate the linear trend component identifying a linear regression trend line interpolating
deseasonalized values across time buckets t ( t = 1, 2, …, T):

6. . Calculate Sales Forecast over the past and the future horizon (multiplicative model): seasonality
(seasonality vector SMt , step 3) multiplied by the trend component (trend line Tt , step 5)

EXAMPLE
Apply multiplicative series decomposition model (MSD) to the quarterly sales of an item, as shown in the
following table
1. Apply a CMA algorithm of step k=4 (to remove seasonality on annual basis)

2/3. Calculate seasonal coefficients for each quarter (mean values to eliminate irregular white noise)

4. Calculate deseasonalized historical series values

5. Calculate the trend line on deseasonalized demand values

NON LINEAR TREND MODELS

INITIALISATION OF HOLT – WINTERS ALGORITHM

Initialize Exponential Smoothing values

How to define initial values for the parameters of single, double, triple exponential smoothing methods? (initial
mean, trend and seasonality values)
Example : given a monthly and seasonal historical series, calculate initial values for mean, trend and
seasonality, analyzing 2 year-historical data

ANALYTICAL METHOD

TREND LINE METHOD

 Comparison between initialization
methods
Different values are calculated as output

of Analytical method and Trend Line method

Rolling forecast
Once initial values of mean, trend and seasonality
have been defined, it is possible to launch the
rolling forecasting calculation, starting from the first future period (in the example: January of Year 3)

INVENTORY PLANNING AND CONTROL

Why to hold inventory ?

1) To absorb demand uncertainty (“value”= product availability)

 Safety stock (to face variability);
 Seasonal stock (to face lack of capacity)
 2) To decouple different operations phases (“value” = supply efficiency)
 To keep stock is a way to allow source, make and delivery systems to work with different rate,
as it allows the different phases to be “asynchronous” and it absorbs upstream variations.
3) To gain profit through speculation (“value” = profitability)

Inventory is…

 a Lever of Efficiency in order to decrease the costs of other processes/parts of the production/logistic
system
 a Lever of Effectiveness, in order to better satisfy the customer needs

Inventory costs

INVENTORY CARRYING COSTS

• Storage costs
o Physical storage cost
o Insurance and Miscellaneous
• Financial assets (i.e. Cost of capital)
• Depreciation
• Obsolescence or spoilage cost

ORDERING COSTS
• Buyer time
• Receiving costs

SUNK COSTS
 inventory can hide technical, organizational and managerial inefficiencies

Inventory level

Reducing the level of inventory (water) allows operations management (the ship) to see the problems in the
operations (the rocks) and work to reduce them

Inventory … to gain Efficiency

• Seasonal-related inventories: they aim

at decoupling a strong seasonal phase
(e.g. market) from a more balanced one
(e.g. production) (tradeoff: production
costs – inventory carrying costs)

• Speculation-related inventories: they aim at leveraging on low price for high quantity purchases
(tradeoff: purchasing costs – inventory carrying costs)
• Distribution-related inventories: they aim at leveraging on low prices for transportation (tradeoff
transportation costs – inventory carrying costs)

Cycle stock - example

• Production anticipated with respect to demand
 • Cycle stock is a decoupling buffer allowing the production system to work with a rate different from the
sales rate

Inventory … to gain Effectiveness

• An order (or line, or case) is filled if all the items ordered are available in the required quantity for the
shipment
• The Fill Rate:
– is correlated with the number and typology of the items kept in the inventory (which part of
the product range) and the quantity held in stock per item
– is directly perceived by the customer if the required order cycle time is zero (e.g.:
supermarkets)
– otherwise influences the average order cycle time: the higher the fill rate the lower the
average cycle time
• Emphasis on the idea of “customer served”

• Useful also for single items measures

 Inventory “effectiveness” (example)

Inventory Turnover Ratio

• It shows how many times the inventory of an item “turns” in the time period (usually it is calculated on
yearly basis)
• The unit of measure is [turns/period] or [1/period]
• Outgoing Flow and Average Inventory Level (AIL) can be measured as preferred: number of pieces,
cases, pallet loads, m2, m3, liters, kg …

Example 1
Item 1 – Sugar Item 2 – Flour

Example 2
Item 1 – Sugar Item 2 – Flour
Days of Supply (coverage period)

 It is the reciprocal of the ITR (=1/ITR)

 It measures the expected time of autonomy (coverage) of the warehouse in case it were not
replenished anymore
 It is equal to the average duration of stay of the goods in the warehouse

Exercise

Inventories in the distribution system

CYCLE STOCK: these inventories deal with the different operative rhythm of two following stages in the supply
chain

SAFETY STOCK: these inventories deal with the uncertainty of both the demand and the replenishment lead
times
IN TRANSIT STOCK: these inventories are in transit between stockings or production points (mainly inside the
vehicles)

Inventory Analysis

What is the “right” level of stock and how to detect the “useless” part?
Tot stock = right cycle stocks + right safety stocks + inefficiency
PARETO’S RULE: INVENTORY

20% of the items (A class) determines 80 % of the value of the stocks

INVENTORY – SALES MATRIX

Attention area
1) CLASS A-A (high sales, high inventory) The items in this cell are
usually a few, so they can be analyzed one by one. They are both
a risk and an opportunity:
i. Stock-outs will cause heavy drops in sales
ii. No other class offers the same opportunities in
terms of inventory reduction

2) CLASS C-A (low sales, high inventory)

• In this cell there are the items that do not produce sales,
but which have a high inventory level, often hard to
reduce
• It is not enough to wait until they are disposed but it is
necessary to force the reduction:
i. Stop the replenishment and and the production
ii. Push the sales through promotions
iii. Alienation

3) CLASS A-C (high sales, low inventory)

• It is a privileged class of items and this can be a best practice of
inventory management to be extended to the other classes
• It can hide some dangers: if the inventory level needs are
underestimated there is a risk of loosing sales
i. check the presence of stock-outs

Possible causes of misalignment of inventory and sales

 Low frequency of replenishment
 Minimum order batch size imposed by the supplier
 Scarce reliability of the suppliers
 Discounts on purchases of large quantities
 Inventory to support New Product launch
 Speculation
 Demand forecast errors
 Not updated inventory management parameters

INVENTORY PLANNING MODELS

FIRST APPROACH: CONTINUOUS REVIEW

Fixed order quantity: The “Economic Order Quantity EOQ – Reorder Point” model
  When: Place an order just when stock level (availability) reaches a given Order Point (OP)  continuous
review
 How much: A fixed quantity is ordered, equal to the Economic Order Quantity (EOQ), which minimizes
the annual total inventory management cost

THE “EOQ - REORDER POINT” MODEL

FIXED ORDER QUANTITY SYSTEM
 HOW MUCH? Place an order just when stock level (availability) reaches a given Order Point
(OP)
 WHEN? A fixed quantity is ordered, equal to Economic Order Quantity (EOQ)

Safety Stock (SS) is useful to prevent uncertainty in demand and in lead-time which could generate
stock-out during the Lead Time

Safety Stock is calculated in order to guarantee that demand variability during LT does not give a stock-out`
Economic Order Quantity EOQ is the lot Q that minimizes annual total cost:

Exercise

Find the lot size that minimizes the total carrying

cost for the given product GR1:

- Purchase price: 5 £/unit

- Carrying cost factor: 30 % / year

- Order cost: 8 £ / order

- Annual demand: 60 units / year

Quantity discounts

In the presence of quantity discounts, the annual total cost contains specified break points:
Step 1: Evaluate the optimal lot size for each price

Step 2: Select the order quantity Q* for each price

Step 3: Calculate the total annual cost of ordering Q* units

Step 4: Select order quantity Q* with the lowest total cost

Exercise

Find the lot size which minimizes the total carrying cost for the given product GR1 :
 Carrying cost factor : 20 % / year
 Order cost : 100 € / order
 Weekly demand : 150 units / week
 Working days : 240 days/year
 Minimum and multiple lot size: 100 units/order
 Purchase price :
o up to 499 units / order 18 € / unit
o between 500 and 999 units / order 16 € / unit
o over 1.000 units / order 12 € / unit

PERIODIC REVIEW
WHEN? Orders are placed every T days (Review Interval)
HOW MUCH ?The order quantity aims at reaching a determined availability target, enough to satisfy the
expected demand of the entire period between two orders (T+LT)

Availability target
 NOTE: the average lot size Q (t) is equal to the average
demand during the time elapsing between two orders
(AVDT ).

Example
Calculate the availability target (AT) level, cycle stock (CS), average inventory level (AIL) assuming that the
periodic review model is adopted, given:

 Replenishment lead time : 3 weeks

 Review interval: 6 weeks
 Average demand : 60 units / year
 Safety Stock : 2 units

UNCERTAINTY MANAGEMENT
If the lead-time and the demand are uncertain, safety stock is required in order to avoid the stock out during
T+LT.

SAFETY STOCKS
Safety Stock is based on the stock cover level
Safety stock faces demand variability during LT and LT variability as well

Safety stock Target

Stock coverage probability

ESTIMATING SAFETY STOCK
Considering a generic product, it’s possible to define the Safety Stock quantity to hold:

Composed standard deviation comprehensive of D variability, concerning average value of LT and LT variability
concerning D average value

HOW TO CALCULATE SAFETY STOCK

Safety stock calculation formula

 SS(t) = safety stock profile for a given SKU (depending on time t)

 K = target service level (ex: 95% probability not to have stock out during a replenishment period)
 σ(t) = forecast error (depending on time t) + supplier reliability

Formula: SS(t) = K σ(t)

HOW TO CALCULATE CUSTOMER SERVICE LEVEL
 Hypothesis: demand during lead time is a random variable normally distributed with mean equal to d
and standard deviation equal to ƠL.
 By defining safety stock level, the probability of stock coverage is univocally defined.

ESTIMATING SAFETY STOCK FOR THE «EOQ MODEL»

DEMAND SYSTEM

If demand and lead time are uncertain variables following a normal

distribution:

You have:

Example:

Stock cover level for item XYZ : 95%

FORECAST
SYSTEM

Same conditions as the demand system but in this case the variability of the demand must be expressed
through the SDE (forecast error) and the value of the average demand must be changed into the forecast
demand value (F)

ESTIMATING SAFETY STOCK FOR THE PERIODIC REVIEW MODEL

Taking into account a generic product, it’s possible to estimate the safety stock quantity to be carried:
 Exercise
Identify the lot size which minimizes total
inventory carrying cost for product GR1
given

Loss integral
Safety stocks - Service Level considerations
We can define the service level as:

 The probability to avoid the stock out during the lead time:

 The item fill rate (IFR): demanded quantity – available quantity ratio

USO = expected annual stock out [units/year]

Dy = expected value of the annual demand

Service Level
Assumption: the demand during the LT can be represented through a normal distribution with average EDLT
standard deviation ƠDLT
Service level - Probability to avoid the stock out
By setting the safety stock level, the probability to avoid the stock out has been set as well

SERVICE LEVEL – ITEM FILL RATE

The Item Fill Rate (IFR) can be calculated as follows
Exercise
Let’s take a store that manages inventory according to a reorder point policy.

The reorder quantity is equal to Q = 60 units

Given:

 Average weekly demand = 240 units/ week

 Standard deviation of the forecast errors = 40 units/week
 Average LT = 10 working days
 LT standard deviation = 3 days

Please determine the safety stock level to ensure IFR > 90%
Economic optimization
Putting the derivative of total cost with respect to SS equal to zero, we obtain probability of not going into
stock-out during the LT (Pr*), which minimizes total costs (maintenance + stockout)

Putting the derivative of total cost with respect to SS equal to zero, we obtain probability of not going into
stock-out during the LT (Pr*), which minimizes total costs (maintenance + stockout)

Example
An item is replenished every 2 weeks with an average lead time equal to 4 days (constant). The weekly average
demand is equal to 4 units and the forecast error SDE is equal to 1.2 units.

Assuming that:

 phc = 30%/year
 soc = 10% of the product price
 Product price P = 30€/unit

You are required to:

 Estimate the SS level that yields the economic optimization (Pr*)

 Assuming a value of IFR = 95%, assess the marginal advantage of increasing the

IFR value to 98% basing your calculations on the yearly total cost (ICC + SOC).

Step 1
First, calculate the probability of not going into stock-out, that minimizes total costs (carrying cost +
stockout cost) = Pr

Being: Lot (Q) = Demand x Reorder Interval = 4 units/ week x 2 weeks = 8 units

Taking a stock coverage value equal to 88.5%, from the table of the normal standard distribution you
get a value of k = 1.2 or in Excel = NORMSINV (0.885) The corresponding safety stocks will be equal to:

Step 2a
To ensure a value of IFR > 95% it is necessary to calculate the value of the coefficient k that
corresponds to the value of the loss integral and to the σ(D,LT)

Step 2b
To ensure a value of IFR > 98% it is necessary to calculate the value of the coefficient k that
corresponds to the value of the loss integral and to the σ(D,LT)
ALLOCATION OF SAFETY STOCK IN THE DISTRIBUTION
NETWORK
SAFETY STOCK IN THE DISTRIBUTION NETWORK

Allocation: assuming that the number of echelons, the number and the capacity of the warehouses, the
location of the plants etc. have already been defined, we can now take into account the different stock
allocation strategies that allow the minimization of the overall cost and the achievement of the required
service level

Where should we keep the safety stock?

SAFETY STOCK IN A 2-ECHELON DISTRIBUTION NETWORK

 COUPLED SYSTEM: Safety stock is allocated
to the RWs (stock in the RWs “depends” on
the production system)
 INDEPENDENT SYSTEM: Safety stock is
allocated to both the RWs and the CW
(stock in the RWs doesn’t depend on the
production system)

COUPLED SYSTEM
 If safety stock is allocated to the Regional Warehouses only, it faces:

The demand variability ƠDi (downstream variability):

 o during the replenishment lead time of the Central Warehouse (LTc)
o during the replenishment lead time of the Regional Warehouses (LTrw)

The lead time variability (upstream variability):

o the replenishment lead time of the Central Warehouse (LTc)

o the replenishment lead time of the Regional Warehouses (LTrw)

INDEPENDENT SYSTEM
If safety stock is allocated to both the Regional Warehouses and the Central Warehouse, it faces:

In the CW the variability of:

• the downstream demand (σDC) during the lead time LTc

• the upstream replenishment Lead Time LTc (i.e. σLTc)

In the RWs the variability of:

 the downstream demand (σDi) during the Lead Time LTrw

 the upstream replenishment Lead Time LTrw (i.e. σLTrwi)

Assumptions

SAFETY STOCKS IN THE COUPLED SYSTEM

The overall safety stock in the system is the sum of the safety stocks in each of the N RW

SAFETY STOCKS IN THE INDIPENDENT SYSTEM

The overall safety stock in the system is the sum of the safety stocks in each RW and the safety stock in the CW
The demand to be fulfilled by the central warehouse is the sum of the demands seen by the RWs

COUPLED SYSTEM VS INDEPENDENT SYSTEM

Inventory allocation in the network - exercise

? SS coupled system

? SS independent system

Number of Warehouses
Assuming that the number of echelons has already been chosen we want to determine the number of
warehouses that minimizes the overall distribution cost for a given service level

How many warehouses?

The main costs in choosing the number of warehouses follow the trend below
N.B.: the relative position of the various curves in the above graph is arbitrary and is actually application-
specific.
REPLENISHMENT PLANNING
STOCK BALANCE EQUATION

REPLENISHMENT & STOCK PROJECTION

Standards of measurement and calculation of values into the stock balance equation:

• Inventory INV(t) is conventionally

projected at the end of each period
t
• Inventory at the end of period t is
equal to the inventory at the
beginning of period t+1
• Replenishment R(t) – production or
distribution volumes – is available
to sell at the beginning of delivery
period t («receipt» date of goods,
first date of inventory holding)
• Demand Forecast F(t), composed by
the mix of forecasts and orders
portfolio, is due and served during the period t, in
a «uniform» shape in the sub periods forming the whole period t

Stock balance equation: INV(t-1) + R(t) = F(t) + INV(t)

 t = January

The standard stock balance equation is not working properly if it happens the following condition:

INV(t-1) + R(t) < F(t)

that is: if the sum of global availability of item volumes is less than the sum of global «demand», thus less than
global amount of resources required in the period t, depending on mix {forecast + orders}.

Example: initial stock INV(t-1) = 100; replenishment R(t) = 20; forecast F(t) = 145

Two possible solutions to the issue of reduced stock availability:

Modeling of stockout L(t) → lost demand

INV(t-1) + R(t) = F(t) – L(t) + INV(t)

Example: .

 ▪ Ending stock INV(t) = 0 → physical stock is zero!

 ▪ Lost demand L(t) = 145 – (100 + 20) = 25
 ▪ Stock balance equation: 100 + 20 = 145 – 25 + 0

The lost demand L(t) = 25 is not subjected to backlog, thus generates stockout, following the MTS (Make to
Stock) «pure» model: the customer demand not served on-time is definitely lost.

Modeling of a delivery delay B(t) → backlog

INV(t-1) + R(t) + B(t) = F(t) + B(t-1) + INV(t)

Example:
 Ending stock INV(t) = 0 → physical stock is zero!
 Backlog demand B(t) → 145 – (100 + 20) = 25
 Stock balance equation: 100 + 20 + 25 = 145 + 0
In the period t, the company has to serve two types of demand:
• the forecast F(t) [on-time customers]
• the backlog B(t-1) cumulated in the previous periods [delayed customers from previous periods t]
→ backlogs can be solved in the next periods

Stock balance equation modification

The standard stock balance equation has to be modified:

INV(t-1) + R(t) = F(t) + INV(t)

adding two components:
• the net requirements coming from safety stock SS(t), to replenish «by delta» (incremental or
decremental)
• the availability of scheduled orders receipts A(t)
INV(t-1) + A(t) + R(t) = F(t) + ΔSS(t) + INV(t) R(t): planned orders receipt
A(t): scheduled orders receipt

Ending inventory INV(t) represents the residual

availability, not used during the period t.

The ending inventory INV(t) is composed by cycle

stock, to be considered distinct from safety stock
SS(t).

FINAL VERSION OF STOCK BALANCE

EQUATION
• B(t): current «delay»,
generated over the forecast
on period t
• B(t-1): total delay,
«cumulated» from previous
periods
• ΔSS(t): planned increment or reduction for the safety stock

DISTRIBUTION REQUIREMENTS PLANNING

Replenishment: terminology
 Demand (gross requirements): Independent demand at store level (consumers)
 Net requirements: Independent demand, nettified by on hand inventory at POS: {Gros Req – Inv} = Net
Req
 Replenishment
o Planned Order Receipts: quantity to send to POS, to cover the net requirements, with a Just In
Time JIT delivery, to avoid stockout (receipts at their physical delivery date)
o Planned Order Releases: quantity to send from the «supplier» node (warehouse) to the
«customer» node (POS) (departure date)  Release Date = {Receipt Date – Distribution Lead
Time}

Example
 Gross Requirement: Forecast = 100 pieces in week w1
 Net Requirement
o Initial stock = 18 pieces at the beginning of week w1
o Lead time = 1 week
o Net Requirement = (100 – 18) = 82 pieces
 Replenishment
o Replenishment rule: multiple lot = 15 pieces
o Replenishment (arriving at the beginning of week w1) = 90 pieces
o Stock at the end of week w1 = 0+18+90-100 = 8 pieces
Basic hypothesis and constraints for INVENTORY & REPLENISHMENT PLANNING

a. Single-level BOM / BOD

 Replenishment of packaged finished products (textile or fashion) @ POS
b. Infinite capacity at the warehouse / supplier (distribution warehouse / plant warehouse)
 Storage capacity @ warehouse
 Production capacity @ plant
c. No alternative sourcing
 Unique supplier / central warehouse for replenishment
d. Delivery logistic constraints (distribution / warehouse)
 Lot sizing (deliveries with minimum, maximum, multiple lot size)
e. Delivery lead time
 The length of supply lead time determines the length of frozen period (first lead time)
f. No mix constraints (multi-item)
 Minimum Supplier Quantity / Minimum Order Quantity
 Assortment mix

DISTRIBUTION REQUIREMENTS PLANNING

Example of textile / fashion constraints related to the product mix

1) Minimum per order (per supplier / period)

 Value constraint: Replenish at least 2.000 € of merchandise (item mix is free)
 Logistic constraint: Replenish at least 33 pallets (1 standard vehicle) of merchandise (item mix
is free)
2) Assortment mix
 Assortment Planning: if you order a quantity Q > 0 of item A, you have also to order at least
10% of Q of item B
 Replenish at least 1 item for each item of a family brand / group (complete mix of colours,
sizes, styles): not only best seller but also slow movers

Project the system of INVENTORY & REPLENISHMENT PLANNING

1. Input parameters definition
2. Replenishment algorithms selection
3. Safety stock calculation formula selection
4. Output KPIs definition (for all periods in the planning horizon)
 Replenishment plan
 Inventory plan
  Expected stockout profile
 Delayed demand profile (backlog)

Replenishment KPI: basic report

Output KPIs

 Replenishment plan in the

planning horizon (Planned
Orders Receipts)
o also Planned Orders
Releases
 Stock projection / profile
(Inventory Plan)
 Stockout profile (lost demand, backlog)

INPUT PARAMETERS
 Lead Time, Review Period
 Frozen plan (past replenishment during the first Lead Time)
 Scheduled orders receipts (schedulati)
 Initial stock & store (on hand)
 Demand (mix forecast / customer orders portfolio)
 Inbound calendar & store (opening days)
 Transportation lot sizes (min, max, multiple)
 Demand Shelf – life (backlog)

REPLENISHMENT POLICIES DEFINITION

 PULL Stock control: Min – Max
 PUSH demand forecasting:
o Fixed lot (EOQ), variable interval
o Fixed interval («stock coverage days»)
o MSQ: Minimum Supplier Quantity
o Assortment mix control: replenishment multi-item and/or multi-site

BASIC RULE: replenishment quantities come to stores always «just in time» (at the latest day), for minimizing
average projected stock profile (consequence of the replenishment algorithm used)

PULL REPLENISHMENT ALGORITHMS

MINIMUM – MAXIMUM algorithm

When the current stock level INV(t) is lower than a minimum MIN, replenish a variable quantity, to reach a
fixed target quantity MAX (maximum stock level)

• when to replenish? Variable interval, if INV(t) < MIN;

• how much to replenish? The variable quantity {MAX – INV(t)}

This roughly simple stock control algorithm does not take into account the future forecast demand. This may
cause stockout in future periods of the planning horizon
REPLENISHMENT & STOCK BALANCE EQUATION
Hypothesis: it is not allowed to have any replenished volume

 Calculate customer service level

 Calculate punctual stock coverage

PUSH REPLENISHMENT ALGORITHMS: ACTIVATION RULE

How to determine the reorder point? Replenishment quantities are requested:
• (fixed quantity: EOQ)
• (variable quantity: to serve forecast into coverage interval)

in a «just in time» way, that is when the total stock availability (total supply), composed by:

is less than the sum of requirements (total demand):

POS INVENTORY & REPLENISHMENT PLANNING

BASIC SCENARIO (NO REPLENISHMENT)
Stock & stockout projection on the future
No replenishment → end of season for a collection item; old items cannibalized by new items; end of stock @
central warehouse during a promotion activity
REPLENISHMENT WITH MIN-MAX ALGORITHM
Min = 25 pieces, Max = 120 pieces; rule: if INV(t) < 25 (Min), Replenish {120-INV(t)} = {Max – INV(t)}

REPLENISHMENT WITH EOQ ALGORITHM (FIXED LOT SIZE)

Economic Order Quantity = 110 pieces; if INV(t) < Demand(t+1), replenish

EOQ = 110 [hypothesis: Lead Time = 1]

REPLENISHMENT WITH COVERAGE PERIODS ALGORITHM (FIXED INTERVAL)
Replenish the quantity needed to cover the total demand in the next 4 periods (starting at the first bucket of
the planning horizon)

Replenish the quantity needed to cover the demand in the next 4 periods, with the additional objective to hold
on stock a fixed safety stock of 18 pieces at the end of each time period in the planning horizon

NUMERIC EXAMPLE
Numeric example: POS 1; LT = 2
Numeric example: POS 2; LT = 2
Numeric example: POS 3; LT = 1; SS = 2
Numeric example: POS 4;
LT = 2; SS = 5

Numeric example:

POS 5; LT = 1; SS = 2