SAP APO SNP Planning Engines Demystified:

Real-World Insights and Best Practices

Mani Suresh
Lennox International
Produced by Wellesley Information Services, LLC, publisher of SAPinsider. 2016 Wellesley Information Services. All rights reserved.

In This Session

Get an overview of the similarities and differences between the various SNP planning
engines and their application to specific supply chain scenarios

Uncover hidden, and often under-utilized, capabilities of each planning engine and get
tips to exploit them to improve planning results

Step through multiple industry use cases for leveraging the SNP optimizer for cost-based
planning, heuristic for infinite planning, and CTM for priority planning

What Well Cover

Supply Chain Management

Supply Planning Approach
Supply Planning with SAP APO
3 SNP Planning Engines and Industry Use Cases
Wrap-Up

What Is Supply Chain Management?

Supply Chain Planning is the design, planning,

execution, control, and monitoring of supply
chain activities with the objective of creating net

value, building a competitive infrastructure,

leveraging worldwide logistics, synchronizing
supply with demand, and measuring

performance globally.

Supply Chain Management

Do I trust my Demand Signal?

Inherently inaccurate in a complex and dynamic market
Biased by sales incentives and promotions

Does my distribution strategy allow the flexibility?

Strategy key to ensuring flexibility
Expedites impacts operating costs and margins
Static methods risk excess inventory positions

Demand

Does my Supply Plan enable me to be successful?

Distribution

Supply

Operating costs directly influence margins and profits

Inaccurate or volatile plan impacts plant efficiency, lost
sales, under-utilized capacity

Supply Chain Management Myths

Costs keep it low

Forecasting accuracy guarantees execution
Technology solves problems bring in the
latest and greatest
Add more capacity add more heads and
hands to it!
5

What Well Cover

Supply Chain Management

Supply Planning Approach
Supply Planning with SAP APO
3 SNP Planning Engines and Industry Use Cases
Wrap-Up

Supply Planning Approach

Robust
Training
and
Education

Business
Process
Transformation

Operational
Master Data
Strategy

Planning
Engine
Selection

Business
Process
Governance

Vision

What problems are we trying

to solve?
What capability are we trying
to add?
What does the long-term
roadmap look like?

Business Process Governance

Measure and baseline

current performance
Ensure the existing business
processes are documented
and adhered to
Business
Process
Governance

Operational Master Data Strategy

Operational
Master Data
Strategy

Are the master data processes

mature enough to handle
continuous planning parameter
changes?

Business
Process
Governance

10

Business Process Transformation

Business
Process
Transformation

Operational
Master Data
Strategy
Business
Process
Governance

Based on the vision

and goals, initiate a
business process reengineering project
This will directly drive
the requirements for
the next phase

11

Planning Engine Selection

Business
Process
Transformation

Operational
Master Data
Strategy
Business
Process
Governance

Planning
Engine
Selection

Only after
these steps,
the planning
engine
selection
makes sense

The decision
is now
directly driven
by business
needs
12

Robust Training and Education

Robust
Training
and
Education

Business
Process
Transformation

Operational
Master Data
Strategy

Planning
Engine
Selection

Business
Process
Governance

Often ignored phase

in change
management
13

Success

Robust
Training
and
Education

Business
Process
Transformation

Operational
Master Data
Strategy

Planning
Engine
Selection

Business
Process
Governance

14

What Well Cover

Supply Chain Management

Supply Planning Approach
Supply Planning with SAP APO
3 SNP Planning Engines and Industry Use Cases
Wrap-Up

15

Supply Planning with SAP APO

Source: SAP

16

Supply Planning with SAP APO (cont.)

Fileserver

Presentation Client is the SAP

GUI End-User Interface.
Application server handles all
APO- and BW-related requests.

Optimizer
Server

Database server contains table

spaces for APO tables and BW
tables.
Optimizer server is a separate
Windows instance to run the
planning engines.
liveCache is an object-oriented
database which is a separate
server in itself.

17

Supply Planning with SAP APO (cont.)

Planning of material flow in the network

Mid- to long-term horizon

DC

DC

DC

Finite bucket-oriented planning

Prioritization of demands and Supply Optimization

18

What Well Cover

Supply Chain Management

Supply Planning Approach
Supply Planning with SAP APO
3 SNP Planning Engines and Industry Use Cases
Wrap-Up

19

Planning Engines in SAP APO

Heuristics

Optimizer

CTM

ABAP-based runs on the

application server
Unconstrained Plan
Has to be followed up by
Capacity leveling
Basic function of
Demand Propagation

C++ Based .
Cost Based Planning.
Finite Planning - entire
network as a single
problem and proposes
the most optimized
solution.
The longest running
engine of the 3.

C++ Based.
Order based finite
planning.
Constraints are taken
into account like
Capacity constraints,
time constraints.

20

Planning Engines in SAP APO (cont.)

Heuristics

Optimizer

CTM

ABAP-based runs on the

application server
Unconstrained Plan
Has to be followed up by
Capacity leveling
Basic function of
Demand Propagation

C++ Based
Cost-Based Planning
Finite Planning entire
network as a single
problem and proposes
the most optimized
solution
The longest-running
engine of the 3

C++ Based.
Order based finite
planning.
Constraints are taken
into account like
Capacity constraints,
time constraints.

21

Planning Engines in SAP APO (cont.)

Heuristics

Optimizer

CTM

ABAP-based runs on the

application server
Unconstrained Plan
Has to be followed up by
Capacity leveling
Basic function of
Demand Propagation

C++ Based
Cost-Based Planning
Finite Planning entire
network as a single
problem and proposes
the most optimized
solution
The longest-running
engine of the 3

C++ Based
Order-based finite
planning
Constraints are taken
into account like
Capacity constraints,
time constraints

22

Planning Engines in SAP APO (cont.)

Optimizer
CTM

Optimized or
optimal Plan
Data-intensive

Feasible Plan
Flexible control
over Planning

Heuristics
Simple demand
propagation
Unconstrained Plan
23

Planning Engines in SAP APO Brief History

Heuristics was the MRP equivalent out of the box

Optimizer was the integration of SAP APO with ILOG CPLEX optimizer a mathematical
solver

CTM was developed for the high-tech industry, semiconductor in particular

24

Planning Engines in SAP APO Common Misconception

Optimizer
is the best!

Heuristics
suffices!
25

Heuristics: Process Flow

Set up Master
data

2
Release Demand
plan to SNP

Run SNP
Heuristics

4
Run Capacity
leveling

Review/adjust
planning results

6
Run
Deployment/TLB

26

Heuristics: Technical Flow

Forecast

Open
Sales
orders

Safety
stock

Calculate Lowlevel Codes

Alerts
Generation

Read planning
data

Capacity
Leveling

InTransit
PO

Stock
on
Hand
Open
PO

Lot size
parameter

Lot size
Rules

Rounding
value

Total
Demand
Total

Total
Supply
Total

Net requirements
Net

Demand

Supply

Requirements

Capacity
Consumption

BOM
Explosion

Plan Order
Creation

Quota
Arrangements

Procurement Type

Transportation lane
lead time

Distribution Demand @ Source

Distribution demand
at Source

Source
Determination

Dependent
Demand for
Components

27

Heuristics: Industry Use Cases CPG, Retail

Large network
Simplistic distribution

Fewer bottlenecks
Make to stock (commodities)

Longer-term planning
What-if scenarios

Less complexity
No shelf-life constraints
Simplistic production process
28

Heuristics: Master Data Requirements

Source of
Supply

Planning
rules

Production
constraints

Transportation lanes lead times

Quota Arrangements
PPM/PDS primary/alternate resource, operations, components

Horizons Production horizon, Forecast horizon, stock transfer horizon

Lot Sizes Min lot, max lot, Lot size strategy, Rounding Values
Safety stock Target stock, Safety stock vs. Safety Days of Supply (SDOS)
Strategy Forecast consumption, Make to Order (MTO), or Make to Stock (MTS)

Resources Capacity, production calendar

Finite/Infinite resource types
Component availability

29

Heuristics: Demand and Supply Components

30

Heuristics: Production Plan

31

Heuristics: Resource Capacity Level in % After Heuristics Run

Peaks and Valleys Characteristic of unconstrained plan
500

450
400
350
300
250
200
150
100
50
0

32

Heuristics: Resource Capacity Level in % After Capacity Leveling

Level loaded constrained plan
140

120

100

80

60

40

20

33

Heuristics: Capacity Leveling Options

Key Decision Factor

34

Heuristics: Comparison of Capacity Leveling Methods

Criteria

Heuristic

Optimizer

Type of processing

Order-based processing:
Only orders that are moved are changed (liveCache, R/3)
Beneficial if there are only a few that are overloading the resource

Period-based and quantity-based processing:

All existing orders are deleted and recreated after capacity
leveling

Resource capacity level

Not always optimal

Almost always optimal, read Perfect!

Changes to plan

Fewer Only where required

Completely new plan

Priority rule observance

Order size, Product priority

No

Overloads

Always in the last bucket

Anywhere in the horizon

Location of receipt and

issue

The receipt has to be within the planning horizon and outside of the
SNP production horizon or stock transfer horizon

The receipt and issue have to be within the planning horizon and
outside of the SNP production or stock transfer horizon

Bucket offset

Is ignored (see period factor)

Period factor

The period factor of the PPM/PDS or transportation lane is taken into

Is ignored (see bucket offset)
account (if not defined, the period factor of the location product is taken
into account)

The bucket offset for PPMs or PDS is always 1

The bucket offset for transportation lanes is always 1

35

Heuristics: Comparison of Capacity Leveling Methods (cont.)

Data set comparisons Example
Optimizer method of leveling
Product

Method

Material 1 b4

201534

201535

201536

201537

120

80

80

60

80

80 120

after

80

201538

80

201539

201540

201541

201542

201543

201544

201545

80

80

80

80

80 100 100

40 200

40

20

60

80

201546

80
180

201547

40

201548

201549

40 120

201550

201551

60 100

201552

60

40 440

Heuristics method of leveling

Product

Method

Material 1 b4 cap lev

After cap
lev

201534 201535 201536 201537 201538 201539 201540 201541 201542 201543 201544 201545 201546 201547 201548 201549 201550 201551 201552

120

80

80

80

80

80

80

80

80

80

100

100

80

40

40

120

60

100

60

120

80

80

80

80

80

80

80

80

80

100

100

80

40

40

120

60

100

100

36

Heuristics: Comparison of Capacity Leveling Methods (cont.)

Changes by category

Optimizer method

Horizons

Till end of 2015

Entire horizon

Unconstrained and Constrained plan

at SKU/Weekly

454/1063 ~43%

1866/3148 ~ 60%

Unconstrained and Constrained plan

at SKU/Buckets(Daily/weekly/
monthly)

NA

NA

Order changes day over day at

SKU/Weekly

711/1014 ~ 70%

1683/2609 ~65%

Order changes at SKU/Buckets

(Daily/weekly/monthly)

907/1253 ~ 72%

1879/2847 ~66%

Shortages

4 SKUs on AL01
37 SKUs on AL02

25 SKUs on AL01
45 SKUs on AL02

Overages

31 SKUs on AL01
36 SKUs on AL02

37 SKUs on AL01
45 SKUs on AL02
37

Heuristics: Comparison of Capacity Leveling Methods (cont.)

Changes by category

Optimizer method

Heuristics Method

Horizons

Till end of 2015

Entire horizon

Till end of 2015

Entire horizon

Unconstrained and Constrained plan

at SKU/Weekly

454/1063 ~43%

1866/3148 ~ 60%

92/992 ~ 9%

239/2606 ~ 9%

Unconstrained and Constrained plan

at SKU/Buckets(Daily/weekly/
monthly)

NA

NA

153/1133 ~ 14%

269/2714 ~ 10%

Order changes day over day at

SKU/Weekly

711/1014 ~ 70%

1683/2609 ~65%

431/1028 ~ 42%

933/2647 ~ 35%

Order changes at SKU/Buckets

(Daily/weekly/monthly)

907/1253 ~ 72%

1879/2847 ~66%

569/1235 ~ 46%

1071/2854 ~38%

Shortages

4 SKUs on AL01
37 SKUs on AL02

25 SKUs on AL01
45 SKUs on AL02

1 SKU on AL01
5 SKU on AL02

1 SKU on AL01
6 SKU on AL02

Overages

31 SKUs on AL01
36 SKUs on AL02

37 SKUs on AL01
45 SKUs on AL02

12 SKU on AL01
19 SKU on AL01

13 SKU on AL01
21 SKU on AL02
38

Heuristics: Capacity Leveling Options

Influence on planning results

39

Heuristics: Process Flow Time Utilization

Time-consuming
activities

Set up Master
data

2
Release Demand
plan to SNP

Run SNP
Heuristics

4
Run Capacity
leveling

Review/adjust
planning results

6
Run
Deployment/TLB

40

Cost Optimizer

C++ based
Runs on the optimizer server
Uses Linear programming/Mixed Integer Linear programming concept
Objective Function
Decision Variables
Constraints
ILOGs CPLEX
SAP IP essentially means a black box
Application of mathematical models

41

Cost Optimizer (cont.)

Objective Function
Business goal that the planner is trying to achieve
What is the problem that is attempted at being solved

Decision Variable
Unknowns impacting the outcome
Demand situation, supply situation
Master data like lot size, lead times, etc.

Constraints
Hard Capacity constraints production/transportation/storage, shelf life, calendar
Soft Capacity constraints Safety stock, due date (demand)

42

Cost Optimizer: Process Flow

Set up Master data

Set up Profiles
Set up Costs

Run SNP Optimizer

Release Demand plan

to SNP

4
Review planning
explanation logs

Adjust planning
results/profiles/costs

43

Cost Optimizer: Technical Flow

Profiles
Costs

Constraints

Read planning
data

Optimization
Optimization Run
run

Plan Order
Creation

Capacity
Consumption

Alerts
Generation

44

Cost Optimizer

Different methods of optimization

Linear
Discrete
Aggregate Plan
Vertical
Horizontal
Decomposition

45

Cost Optimizer: SNP Optimizer

SNP Optimizer

Profiles
Optimizer profile
Cost profile
Optimization-Bound profile
Penalty Group profile
Priority Profile
Planning profile
Lot size profile
Demand profile
Supply profile
Parallel processing profile
46

Cost Optimizer: SNP Optimizer (cont.)

SNP Optimizer

Profiles

Costs

Optimizer profile
Cost profile
Optimization-Bound profile
Penalty Group profile
Priority Profile
Planning profile
Lot size profile
Demand profile
Supply profile
Parallel processing profile

Penalty Costs:
Unfulfilled Demand or delay
Safety stock penalty
Control costs:
Storage Costs
Production
Transportation
Resource Expansion
Procurement
47

Cost Optimizer: SNP Optimizer (cont.)

SNP Optimizer

Profiles

Costs

Constraints

Optimizer profile
Cost profile
Optimization-Bound profile
Penalty Group profile
Priority Profile
Planning profile
Lot size profile
Demand profile
Supply profile
Parallel processing profile

Penalty Costs:
Unfulfilled Demand or delay
Safety stock penalty

Hard Constraints:
Capacities
Shelf life
Material availabilities
Calendar
Duration of operations

Control costs:
Storage Costs
Production
Transportation
Resource Expansion
Procurement

Soft Constraints:
Safety stock
Due dates (demand)
48

Cost Optimizer: Profiles

Profiles:
Profile

Use

Optimizer profile

Define Linear programming method based on constraints

Cost profile

Prioritize different costs with cost weighting factors

SNP Optimization-bound profile

Benchmark with previous runs

Limit the changes in the short term but allow medium/long term.

SNP lot size profile

Maintained in transportation lane

SNP penalty cost group profile

Works with descriptive characteristics

Prioritize different demands based on penalty cost

SNP priority profile

Define product and resource decomposition

SNP planning profile

Maintain in global parameter settings

Demand profile

Maintained in material master to determine the demands

Supply profile

Maintained in material master to determine the supplies

Parallel processing profile

Technical profile used to define parameters for breaking the run into parallel processes
49

Cost Optimizer: Costs

Type

Cost

Impact

Control Cost Higher relative storage cost

Moves products to a location with lower cost irrespective of requirement dates

Control Cost Transportation costs

Prioritize source location

Control Cost Production cost

Priority of manufacturing location

Control Cost Storage costs

Deplete inventory before new build

Penalty Cost Higher non-delivery penalty

Force production

Penalty Cost Higher delay costs

Controlling lateness/pre-build

Penalty Cost Safety stock violation penalty cost

Meeting inventory target

50

Cost Optimizer: Master Data Requirements

Source of
Supply

Planning
rules

Production
constraints

Transportation lanes lead times

PPM/PDS primary/alternate resource, operations, components, scrap, resource consumption

Horizons Production horizon, Forecast horizon, stock transfer horizon

Lot Sizes Min lot, max lot, Lot size strategy, Rounding Values
Costs Safety stock violation penalty costs, Demand violation costs, procurement costs, cost multipliers
Strategy Forecast consumption, Make to Order (MTO), or Make to Stock (MTS)

Capacities and Costs Production, Transportation, Handling, Storage

Shelf life

51

Cost Optimizer: Process Flow Time Utilization

Set up Master data

Set up Profiles
Set up Costs

Run SNP Optimizer

Release Demand plan

to SNP

4
Review planning
explanation logs

Adjust planning
results/profiles/costs

52

Cost Optimizer: Industry Use Cases Aerospace, Oil, and Gas,

Pharma

Capital-intensive
Every step cost is a constraint

Higher complexity
Essential to factor in all the constraints and complexity

Shelf life
Redeploy vs. additional production

53

Cost Optimizer: Profile Screen

Selection assignments

Various profile assignments

Quota arrangements calculated by

optimizer for subsequent use in
other planning engines.
54

Cost Optimizer: Automatic Cost Model

Alternative to the traditional cost maintenance

Uses
Business logic
Demand class vs. location products
Priorities and preferences to determine cost factors
T-lane priority
PPM/PDS priority
Demand priority classes
Location product priority
Product priority

55

Cost Optimizer: Deployment Optimizer

To create a distribution plan considering

Distribution rules such as push rules and fair share
Costs defined in the model such as transportation, storage costs, and non-delivery
penalty costs
Constraints such as transportation capacity, storage capacity, and lot sizes
Different results from SNP optimizer
Focuses on distribution and does not factor in additional supply

56

Cost Optimizer: Connecting to External Optimizer

Connections possible through BAPIs

Integrate through Extension workbench (APX)
Ability to call the external optimizer directly from SAP APO
Control of communication between the optimizer and front end
Transfer of data like it does to the ILOG CPLEX
Extensions using BAdIs

57

CTM

Multi-level supply demand matching

Order-based planning
Prioritize Demand and Supply
Rules-based planning
Finite planning
Pegging results trace back supply to demand
Results in a feasible plan may not be optimal

58

CTM: Process Flow

Set up Master
data
Set up CTM
planning profile

Run CTM

Release
Demand plan to
SNP

Run TLB

Review/adjust
planning results

59

CTM: Technical Flow

Forecast

Open
Sales
orders

Safety
stock

Check Master
Data

Alerts and
Explanation Log
Generation

Read planning
data

Early/Late demand
fulfillment

Total
Demand
Demand

Prioritization

Capacity
Consumption

InTransit
PO

Stock
on
Hand

Lot size
parameter

Open
PO

Lot size
Rules

Search
Strategy

Net requirements
Supply
Demand
Matching

BOM
Explosion

Procurem
ent Type

Transport
ation lane
lead time

Rounding
value

Total
Supply
Supply

Quota
Arrangem
ents

Plan Order
Creation

Distribution Demand @ Source

Multi-source/level Supply
Demand Matching

Source
Determination

Dependent
Demand for
Components

60

CTM: Industry Use Cases High-Tech and Manufacturing

Feasible plan
To generate a feasible plan considering all the constraints from the get go
Evolve this feasible plan as the distribution demand position changes

Large number of products

With different work content
Complexity of setups, multiple operations, and inter-operation intervals

Prioritization
Customer prioritization
Order prioritization Customer order vs. forecast or targets
61

CTM: Planning Scope

Ability to pick and choose by order type

62

CTM: Planning Strategies

Excellent control over planning strategies

63

CTM: Aggregation Techniques

64

CTM: Aggregation Techniques (cont.)

Aggregation options on top of order-based planning

65

CTM: Demand Prioritization

Detailed control over prioritization of demand

66

CTM: Supply Strategy

Determine how to consume the supply

67

CTM: Additional Planning Settings

Further flexibility to control

publication
Control on source determination

68

CTM: Master Data Settings

Flexibility over master data usage

Controls max duration between 2

operations

69

CTM: Technical Settings

Speed up planning runs.

However, does not work with
interchangeability or substitution rules
Explanation logs

70

CTM: Control Parameters

Additional flexibility to control the planning behavior

71

CTM: Global Settings

72

CTM: Global Settings (cont.)

BAdI /SAPAPO/CTM_PLPAR Change CTM global parameters

73

CTM: BAdIs

/SAPAPO/CTM_MCHECK Master data check

/SAPAPO/CTM_ORDERS Modify orders before save
/SAPAPO/CTM_MATLOC Material location attribute change before planning
/SAPAPO/CTM_SUBST Substitution rules change
/SAPAPO/CTM_SOSEXT External source of supply determination
/SAPAPO/CTM_SOSINT Internal source of supply determination
/SAPAPO/CTM_RESOURCE Resource attribute modification
/SAPAPO/CTM_PLPAR Change CTM global parameters

74

CTM: Process Flow

Set up Master
data
Set up CTM
planning profile

Run CTM

Release
Demand plan to
SNP

Run TLB

Review/adjust
planning results

75

3 SNP Planning Engines Comparisons

Heuristics

CTM

Optimizer

Unconstrained

Feasible

Optimized

None

Demand, Supply

All Constraints by cost

Master Data

Low

Moderate

High

Flexibility

Low

High

Moderate

Results Explanation

Simple

Complex

Very Complex

Ease of implementation

Simple

Moderate

Complex

Ongoing maintenance

Less

Moderate

Lot

Long-Range Planning

Feasible Plans

Bottleneck Optimization

Speed
Quality of Plan
Prioritization
Multi-level

Use case

76

3 SNP Planning Engines Total Cost of Ownership

One-Time Cost

Optimizer

CTM

Heuristics

Continual/Recurring Cost

77

3 SNP Planning Engines Key Considerations

Standalone solution
Heuristics CPG
CTM High-Tech
Optimizer Aerospace

Combination solution
Heuristics + CTM High-Tech
CTM + optimizer Aerospace/Process Industry

78

Performance Monitor: Managing Planning Engine Performance

Transaction: /SAPAPO/PERFMON
Statistics for optimizer runs
Runtime comparisons
Monitor runtime variations for different methods of optimizations
Breakdown of runtime by steps

79

What Well Cover

Supply Chain Management

Supply Planning Approach
Supply Planning with SAP APO
3 SNP Planning Engines and Industry Use Cases
Wrap-Up

80

Where to Find More Information

Sandeep Pradhan, Demand and Supply Planning with SAP APO (SAP PRESS, 2012).
Balaji Gaddam, Capable to Match (CTM) with SAP APO (SAP PRESS, 2009).
Jochen Balla, Frank Layer, Production planning with SAP APO (SAP PRESS, 2015).
http://help.sap.com
Follow Supply Chain Management SAP Supply Chain Management SAP SCM 7.0
EHP3 SAP Library SAP Advanced Planning and Optimization (SAP APO)
Supply Network Planning
Follow Supply Chain Management SAP Supply Chain Management SAP SCM 7.0
EHP3 SAP Library SAP Advanced Planning and Optimization (SAP APO)
Multilevel Supply and Demand Matching (SDM)

81

7 Key Points to Take Home

Determine your strategic goals

What problems are you trying to solve?
What capability are you trying to add?
Supply chain (production and distribution) complexity
Measure baseline performance
Consider total cost of ownership, including time and effort and capital and expenses
Remember, one size does not fit all in supply chain
A combination of planning engines works better in certain cases
Track KPIs and focus on continuous improvement

82

Your Turn!

How to contact me:

Mani Suresh
Email: [email protected]
Twitter: @shsmani

Please remember to complete your session evaluation

83

Disclaimer
SAP and other SAP products and services mentioned herein as well as their respective logos are trademarks or registered trademarks of SAP SE (or an SAP affiliate company) in Germany and other
countries. All other product and service names mentioned are the trademarks of their respective companies. Wellesley Information Services is neither owned nor controlled by SAP SE.

84

Wellesley Information Services, 20 Carematrix Drive, Dedham, MA 02026

Copyright 2016 Wellesley Information Services. All rights reserved.