Download - Eastlog 2014 prague
www.jaguar-aps.com
®
Copyright: Jaguar-APS
Effective Use of Data to Support Successful Supply Chains.
Charles Novak & Pavel Černý, Jaguar-APS
Prague, CZ May 2014
www.jaguar-aps.com
Copyright: Jaguar-APS
Theme of the Presentation
Click icon to add picture
®
Copyright: Jaguar-APS 3®
Copyright: Jaguar-APS
Complexities Affecting Business Strategy Factors impacting supply
chain complexity: Globalization Mass customization of
products Compressed product
lifecycles Outsourcing New product introductions Competitive market
entrants Fluctuation in supplier
costs Capacity constraints Mergers and acquisitions
4Copyright: Jaguar-APS
What can we do to avoid this situation…
Utilize and work with all available data!
®
Copyright: Jaguar-APS 5®
Copyright: Jaguar-APS
Managing by Analytics Analytics resolve
differences of opinion. Initial discussion based
on opinions but has to be supported by numbers.
Cross-functional involvement to improve alignment.
Typically, managing by analytics is a MAJOR CHANGE.
Teams work best when analytics rule discussion
Analytics is Much More Than Reporting
Traditional supply chains respond, but they do not sense.
Organizations do not use analytics to listen.
In the world of new analytics, they will be able to sense, test and learn and orchestrate the response market to market.
®
Copyright: Jaguar-APS 6®
Copyright: Jaguar-APS
Analytics is Much More Than Reporting New technology forms
of analytics are expanding the possibilities.
The evolution of analytics for visualization, pattern recognition, unstructured text mining and parallel processing are converging to drive a new form of supply chain.
Think
Sense
Act
Analytics combines digital with cognitive reasoning to sense, think and act.
®
Copyright: Jaguar-APS 7®
Copyright: Jaguar-APS
Analytics is Much More Than Reporting What if we could test and
learn in-market, reading market impacts in real-time through analytics, based on matching customer attributes to product attributes to build customized products for regions around the world?
This new approach allows test and learn capabilities to answer the questions that we do not know to ask to build unique insights.
And, what if we could mine unstructured data and combine it with transactional data to mitigate supplier risk?
www.jaguar-aps.com
Copyright: Jaguar-APS
Business FailureLinking strategy with execution is the key!
Focus on decisions rather than the information.
The reason why successful companies fail is they invest in things that provide the most immediate and tangible evidence of achievement.”
®
Copyright: Jaguar-APS 9®
Copyright: Jaguar-APS
The Roadmap to Sustainable Success Identify business goals Identify business decisions
driving the goals Identify people that will
drive the business goals Identify business processes
that will help with making decisions
Identify data and tools required to make the decisions
Improve planning by launching what-if capabilities.
Strategy
Data-Tools
Supply chain must identify business
goals from the start.
www.jaguar-aps.com
Copyright: Jaguar-APS
The only thing that is constant today, is the CHANGE
Click icon to add picture
®
Copyright: Jaguar-APS 11®
Copyright: Jaguar-APS
Role of Demand Planning Analytical approach
incorporating domain knowledge, not just judgement.
Non-biased forecast generation.
Forecast model and methods evaluation and selection
To tell the top management the
truth!
Data collection, coordination, transformation and analysis
Forecast development and analysis Forecast communication Error tracking, analysis and remediation Cross-functional interface Planning process interface
®
Copyright: Jaguar-APS 12®
Copyright: Jaguar-APS
Introduction: Recent Developments
Predictive analytics used to:
• Uncover patterns in consumer behavior.• Measure effectiveness of marketing investment.• Optimize financial performance.• Shape and proactively drive demand using what-if simulations.• Sense demand signals and shape the future demand supported by data
mining technologies.
®
Copyright: Jaguar-APS 13®
Copyright: Jaguar-APS
Predictive Analytics and Supply Chains• Cost perceived too high• Focus on investments that provide immediate and
tangible results.• Too complex to connect the data nodes across
extended supply chain.• Big data is a distraction at the moment. • Skillsets in supply chain and IT are limited.• Disconnect between the need and silo managed IT and
other departments.• Demand planning not a Core Competence.Why Not• Strong potential to transform the way in which supply
chain managers lead and supply chains operate.• Senior management’s need to grow the business
profitably. • Multi-echelon supply chains require quick and correct
signals to operate effectively. • “Heads-up” to help sense, analyze, and better respond
to market changes.• Data equals information and information equals profit.• Pressures to synchronize demand and supply to
understand why consumers buy products.Why Yes
®
Copyright: Jaguar-APS 14®
Copyright: Jaguar-APS
Demand-Driven Forecasting & Supply Chain
Demand Sensing
Source: Laura Cecere, Supply Chain Insights & Charles Chase, SAS
®
Copyright: Jaguar-APS 15®
Copyright: Jaguar-APS
Demand-Driven Forecasting & Supply Chain Demand sensing
Shortening the time to sense “true” market data to understand “true” market shifts in the demand response.
Order-to-shipment data that can have 1-3 weeks latency in translating “true” market (or channel) demand to action.
Development of Consumption based Forecast
Demand shaping Domain knowledge vs. judgement. Applying techniques to stimulate
market demand. Linking demand to supply chain. Elimination of politics and bias in
forecast. Demand translation
Translating demand outside-in from the market to each role within the organization
The system design recognizes that the requirements for each - distribution, manufacturing and procurement - are different.
In this process, the forecast is based on “the selling unit” into the channel with “ship-to modeling.”
The demand is then translated to “ship-from” views based on the needs of the specific role.
Demand orchestration and shifting Making trade-offs market-to-
market based on the right balance of demand risk and opportunity.
These trade-off decisions depend on the use of advanced analytics to sense and shape demand simultaneously.
Domain knowledge vs. judgment.Source: Laura Cecere, Supply Chain Insights & Charles
Chase, SAS
www.jaguar-aps.com
Copyright: Jaguar-APS
Things You Need To Know About The Process
Process is more
important than
forecasting models and forecasting software.
17Copyright: Jaguar-APS
Demand-Driven Forecasting Process
Source: Charles Chase, SAS
Demand Sensing
Sensing demand signals related to the marketplace by market, channel, category, and product.
Demand Shaping
Using what-if analysis, demand planners shape future demand based on tactical and strategic sales/marketing plans.
Demand Shifting
Collaboration with sales, marketing and operations planning to match supply to unconstrained demand.
Demand Response
Constrained demand response used to develop a final supply response (plan).
Process Flow
Uncover Market
Opportunities and Key Business Drivers
• Sales• Marketin
g
Optimize Sales and Marketing Tactics and Strategies
•Sales•Marketing
Assess Financial Impact
•Finance
Finalize Unconstrained Demand
Forecast
•Sales•Marketing
Consensus Planning Meeting
•Sales•Marketing
•Finance•Operations Planning
Rough Cut Capacity Planning Review
•Operations Planning
Revised Demand Response
•Sales•Marketing
Create Supply
Response
•Operations Planning
Collaborative Work
Flow
Sales and Operations Planning
Consensus Planning
Horizontal Processes
®
Copyright: Jaguar-APS 18®
Copyright: Jaguar-APS
Demand-Driven Forecasting Planning Process This structured approach puts the burden of
ACCOUNTABILITY on the SALES AND MARKETING organizations to produce a more accurate unconstrained demand forecast that reflects current market conditions – assuming there is unlimited supply.
This change in the process signals a radical shift in the way companies view their demand forecasting and planning process today. Most demand forecasting processes are SUPPLY DRIVEN with little emphasis on predicting unconstrained demand, let alone shaping future demand.
Demand driven forecasting is a proactive structured process that senses demand signals and shapes future demand based on sales and marketing strategies and tactics rather that reacting to past supply constraints.
Source: Charles Chase, SAS
®
Copyright: Jaguar-APS 19®
Copyright: Jaguar-APS
Demand-Driven Forecasting Planning Process Focus on demand-driven including evaluation of
strategic, operational, and tactical plans to consolidate departmental inputs by identifying, assessing, and closing any financial gaps.
Provides realistic view of true unconstrained demand.
Based on sales and marketing consensus that has been financially analyzed for further refinement to be used in support of the S&OP process.
Supply constraints are addressed through shifting demand to allow for efficient and cost-effective response to meet demand.
Source: Charles Chase, SAS
®
Copyright: Jaguar-APS 20®
Copyright: Jaguar-APS
Vendor
DC
Store
Store
Store
Inside-out (push) strategy issues
Downstream demand accumulated ad presented as aggregate total.
Delay in the initial demand from original customer.
Service level need is an average.
Upstream supply expected to be at 100% service level.
“If replenishment takes care of inventory problems, what caused the inventory problems in the first place?”
Inventory
Information
®
Copyright: Jaguar-APS 21®
Copyright: Jaguar-APS
Inside-Out Questions Is the demand the result of consumption at a
store level? Is the demand upstream the result of the
downstream consumer demand or the downstream warehouse demand?
How long it is before a forecast becomes true demand?
Does the demand signal become true just because you believe it to be better?
Do you have enough time to react to the true demand when it finally becomes known?
Planning
Production Sales
PUSH
®
Copyright: Jaguar-APS 22®
Copyright: Jaguar-APS
More intelligent use of market data Changing view of how to
look at the data and serve customers From INSIDE-OUT to
OUTSIDE-IN It is about owning the entire
supply chain, including the channel, and managing products from their manufacture to their end use.
Current supply chains catch orders and shipments and assume that they are representative of the market.
Supply chain leaders can sense dynamics in the market and translate it through organization and alert the right person at the right time to make the right next action.
Planning
Production Sales
PULL
23Copyright: Jaguar-APS
Inside-Out (PUSH) Outside-In (PULL)
Strategic Focus
Superior new products Product solutions to customer’s unmet needs
Investment Required
High; requires engineering innovation
Small to high; requires marketing brains & insight in consumer behavior
Marketing Communication
Product features and benefits
Solutions to customer problems
Measurement
Sales & market share Customers share & lifetime value
Profitability Short term planning; profits subject to market changes
Long term planning; profits subject to customer loyalty
Planning Demand and supply based on past shipments/orders
Translating market demand to each role within the organization.
®
Copyright: Jaguar-APS 24®
Copyright: Jaguar-APS
Infrastructure for Demand-Driven Value Networks
Opportunities
Insights RiskDemand Supply
Become Market Driven
SustainabilityBuild Value in
Supply Networks
Drive Innovation in Products and Services
Orchestrate Demand-Driven
Response
Demand-driven value networks (DDVNs) integrate processes and data in the supply chain to enable collaboration, as well as orchestrate a response to demand that creates value and mitigates risk.
Source: Gartner Research
®
Copyright: Jaguar-APS 25®
Copyright: Jaguar-APS
Demand-Driven Value Network
Focus of DDVN
Process
Network Design
Sensing
Proven Benefits: Improved customer
service Lower inventory levels Improved working
capital and cash flow Reduced supply chain
disruptions Improved operational
efficiency Lower IT costs Improved profitability Improved market share
Source: Gartner Research
®
Copyright: Jaguar-APS 26®
Copyright: Jaguar-APS
Supply chain managers must advocate for “demand-shaping with supply in mind” Price
These activities cause significant demand variation and uncertainty depending upon the price elasticity of products and competitive reactions.
Supply chain managers can support pricing decisions and demand variations and uncertainties.
For example, they can carry material and component buffer stocks, reserve excess manufacturing capacity, and maintain safety stocks of finished goods.
Promotion As with pricing activities, these cause significant demand
variations and uncertainties. Prior to a promotional campaign the primary role of supply
chain managers is to fill downstream supply chains with product to cover the often substantial uplift in expected demand.
Larry Lapide, MIT Supply Chain Management
®
Copyright: Jaguar-APS 27®
Copyright: Jaguar-APS
Supply chain managers must advocate for “demand-shaping with supply in mind” Product
Establishing and changing the portfolio of products sold, including the introduction of new and reformulated products and the phasing out of old obsolete products.
New product launches especially have significant demand uncertainty. Yet it is important to ensure that product is available to satisfy first-time buyers.
Supply chain managers need to execute launches by initially filling downstream supply chains with sufficient inventories, as well as helping to ensure new products are positioned at the points of sale.
As a product launch progresses, supply needs to be replenished all along the downstream supply chains, as well as at the points of sale.
Place Involves establishing the distribution and sales channels. Similar to new product launches, opening a new channel involves very significant
demand variation and uncertainty. It involves establishing the ways products will flow and be inventoried throughout
a new channel, as well as initially stuffing and replenishing it with inventory. For example, establishing an online Internet sales channel often involves
deploying new order fulfillment and supply strategies, such as piece picking, packing, and shipping in customer-facing warehouses.
Larry Lapide, MIT Supply Chain Management
www.jaguar-aps.com
Copyright: Jaguar-APS
Data segmentationStatistical Methods Selection Based on Segmentation and Portfolio Management
Click icon to add picture
𝐷𝑒𝑚𝑎𝑛𝑑 𝑃𝑎𝑡𝑡𝑒𝑟𝑛𝑠+𝐷𝑒𝑚𝑎𝑛𝑑𝑉𝑜𝑙𝑎𝑡𝑖𝑙𝑖𝑡𝑦=𝐹𝑜𝑟𝑒𝑐𝑎𝑠𝑡𝑎𝑏𝑖𝑙𝑖𝑡𝑦
Source: Charles Chase, SAS
Data
Trend, Seasonality,
Cycle, Randomness
High Value, Low Value
New, Harvest, Growth, Niche
Forecastability
Value to the Company
Data Segmentation
New Products
Niche Brands
Growth Brands
Harvest Brands
®
Copyright: Jaguar-APS 29Copyright: Jaguar-APS
Four Quadrants Based on Portfolio Management
Source: Charles Chase, SAS
Low ValueHigh
Forecastability
Low Priority Products:· Strong Trend· Highly Seasonal· Possibly Cycles· Minor Sales Promotions
Low Priority Regional Specialty Products:
· Some Trend· Seasonal Fluctuations· Irregular Demand· Local Targeted Marketing
Events
Product Line Extensions:(Evolutionary New Products).Some ‘Like’ history available.
Short Life Cycle Products:Many ‘Like’ products available.
New Products:(Revolutionary New Products)
No ‘Like’ history available.
High Priority Products:· Strong Trend· Seasonal Fluctuations· Possible Cycles· Sales Promotions· National Marketing Events· Advertising Driven· Highly Competitive
Growth Brands
Niche Brands Harvest Brands
New ProductsCo
mpa
ny V
alue
Forecastability
High ValueLow
Forecastability
Low ValueLow
Forecastability
High ValueHigh
Forecastability
Low ValueHigh
Forecastability
®
Copyright: Jaguar-APS 30Copyright: Jaguar-APS
Statistical Methods Selection Based on Segmentation and Portfolio Management
Source: Charles Chase, SAS
ARIMAXARIMA with Interventions and Regressors
Simple RegressionMultiple Regression
Combined average:Judgment, Time Series, Causal
Combined Weighted:Judgment, Time Series, Causal
Croston’s Intermittent Demand
ARIMA Box-JenkinsWinters 2 / 3 ParameterDecompositionSimple Moving AverageHolt’s Double Exponential Smoothing
‘Juries’ of Executive OpinionDelphi Committees
Sales Force CompositesIndependent Judgment
Causal Modeling
Multiple Methods Time Series
JudgmentalCo
mpa
ny V
alue
Forecastability
New Products
High ValueLow
Forecastability
Niche Brands
Low ValueLow
Forecastability
Growth Brands
High ValueHigh
Forecastability
Harvest Brands
Low ValueHigh
Forecastability
®
Copyright: Jaguar-APS 31Copyright: Jaguar-APS
Data Segmentation for Effective Demand Planning
Sample size: 14,271 SKUs
®
Copyright: Jaguar-APS 32Copyright: Jaguar-APS
Data Segmentation - Forecastability
www.jaguar-aps.com
Copyright: Jaguar-APS
Collaboration accross the supply chainWhat can you get from your customers or what can you get from your suppliers.
®
Copyright: Jaguar-APS 34Copyright: Jaguar-APS
Bullwhip effect:Increasing Variability of Orders Up the Supply Chain
Bullwhip effect signals less than optimal supply chain caused by inefficiencies in information sharing. Poorly managed supply chains cause instabilities (i.e. poorly forecasted consumer demands). These instabilities create these bullwhip effect swings.
®
Copyright: Jaguar-APS 35®
Copyright: Jaguar-APS
Solutions to fixing the Bullwhip effect Sharing of real demand data across the supply
chain Trust and collaboration – process integration Agile execution to respond to demand
variability Use “Pull” – Outside-in techniques
Customer demand drives material flow Consumption based replenishment and forecasting Frequent deliveries in smaller volumes to replace
customer consumption VMI CPFR
36Copyright: Jaguar-APS
Vendor Managed Inventory VMI
Collaboration Process of Manufacturer and Distributor/Retailer
Optimization of Supply Chain performance
Manufacturer responsible for maintaining the levels of inventory at Distributor’s location(s)
Manufacturer has access to Distributor’s inventory data and generates POs
37Copyright: Jaguar-APS
Collaborative Planning, Forecasting and Replenishment (CPFR)
Communication and Sharing of Demand Plans with External Partners.
CPFR combines the intelligence of multiple trading partners in the planning and fulfillment of customer demand.
Links sales and marketing best practices to supply chain planning and execution processes.
Objective: Increase availability to the customer. Reduce inventory, transportation and
logistics costs.
®
Copyright: Jaguar-APS 38Copyright: Jaguar-APS
Collaborative Planning, Forecasting and Replenishment
PUSH
ConsumptionShipments = 31%
more than consumption
PULL
Consumption
Shipments = 98% of consumption
Before CPFR
With CPFR
®
Copyright: Jaguar-APS 39®
Copyright: Jaguar-APS
Questions that are answered through collaboration with customers / distributors What is my company
impact on my sales due to retail price?
Which of my items are on each planogram?
What stores are not selling my key items?
How do various store groups/regions perform with my product portfolio?
Which store has which planogram?
What are the top 20% of stores contributing to 80% of my business loss?
What are my SKUs performing at this retailer and where should I look for distribution gains?
How do my SKUs perform on ad and various price points?
What were my sales last week? Last month?
Are my innovation SKUs getting to shelf edge quickly?
How do my sales compare to previous timeframes?
®
Copyright: Jaguar-APS 40®
Copyright: Jaguar-APS
Vendor Reporting
Stores not selling SKU rank Speed to market Planogram crosstab
by store Planogram crosstab
by item Store performance
report
Vendor scorecard Pricing and
promotional analysis Store cluster
comparison Performance trends Weekly sales raw data
Types of reports that enhance Domain Knowledge and support Demand Sensing and Shaping process.
What information you can get from your distributors or customers? What
information can you share upstream in your supply chain?
www.jaguar-aps.com
Copyright: Jaguar-APS
Forecasting Outside-inLinking Market Data to Shipments – Simplified Example
®
Copyright: Jaguar-APS 42®
Copyright: Jaguar-APS
MTCA – Multi-Tiered Causal Analysis
Past constraints are becoming non-issue today: Data collection and storage Computing power available Data synchronization capabilities Analytical expertise
MTCA, a process of nesting causal models together using data and analytics, considers marketing and replenishment strategies jointly, rather than creating two separate forecasts.
“Integrating consumer demand into the demand forecasting process to improve shipment (supply) forecasts has become a high priority in the FMCG/CPG industry as well as in many other industries over the past several years.” Charles Chase, SAS
Source: Charles Chase, SAS
®
Copyright: Jaguar-APS 43Copyright: Jaguar-APS
Growth brand – traditional forecasting approach
What do you think? Pretty good forecast – isn't it?Not so fast …
®
Copyright: Jaguar-APS 44Copyright: Jaguar-APS
MTCA – Multi-Tiered Causal Analysis
Market data from ACNielsenFactory shipments
®
Copyright: Jaguar-APS 45Copyright: Jaguar-APS
1. Data Analysis
Comparisson of category consumption and forecast with all outlet consumption for brand. Strong correlation between the two variables is confirming origninal assumptions of category influencing brand consumption.
®
Copyright: Jaguar-APS 46Copyright: Jaguar-APS
2. Development of Consumption Forecast
Regression 1: Last 4 Periods Out of Sampleb 0.670646293 114991.935 a
SE X 0.01659508 43476.00362 SE YR2 0.989098563 57723.81787 SE
F 1633.158501 18 df Residual
SS Reg 5.44175E+12 59976704694 SS Residual
t 40.41235579
Out of SampleForecast Bias % Bias Abs Error APE4,247,556 329,111 7% 329111.089 7%1,997,206 (131,092) -7% 131091.571 7%1,772,187 (44,406) -3% 44405.9146 3%1,724,543 (11,602) -1% 11601.8607 1%
ME -1% MAPE 4%
Regression 2: All Data In Sampleb 0.716448119 -987.7474012 a
SE X 0.015630716 44552.36932 SE YR2 0.989636968 78137.09794 SE
F 2100.930716 22 df Residual
SS Reg 1.2827E+13 1.34319E+11 SS Residual
t 45.83591076
Running two regressions: first to validate the model, second to use the model to generate forecast for the brand‘s consumption.
1
2
®
Copyright: Jaguar-APS 47Copyright: Jaguar-APS
3. Development of Shipment Forecast Based on Consumption ForecastDetection of lag/lead relationship of
factory shipments and consumption.
Shipment forecast based on consumption.
®
Copyright: Jaguar-APS 48Copyright: Jaguar-APS
4. Linking Consumption Forecast to Supply Chain and Internal Marketing/Sales Programs
Adding supply events and TV advertising (dummy variables) plus marketing promotions (past history and forecast) as final variables to the consumption based factory shipment forecast.
Final forecast (green) based on consumption, supply chain constraints, marketing and sales activity.
®
Copyright: Jaguar-APS 49Copyright: Jaguar-APS
Comparison of traditional forecasting approach (blue) versus MTCA (red).
What would it mean to supply chain if only statistical shipment forecast was used?
®
Copyright: Jaguar-APS 50®
Copyright: Jaguar-APS
Why Haven’t Companies Embraced the Concept of Demand-Driven?
IncentivesTraditional view of supply chain excellence
Leadership Focus: Inside out, not outside in
Vertical rewards versus horizontal processes
Focus on transactions not relationships
Source: Charles Chase, SAS
®
Copyright: Jaguar-APS 51®
Copyright: Jaguar-APS
Why Haven’t Companies Embraced the Concept of Demand-Driven? Incentives:
As long as sales is incented only for volume sold and marketing only for market share, companies will never become demand driven. To make the transition to demand-driven, companies must focus on profitable sales growth through the channel.
Traditional view of supply chain excellence. For demand-driven initiatives to succeed, they must extend from the customer's customer to supplier’s
supplier. Customer and supplier initiatives usually are managed in separate initiatives largely driven by cost.
Leadership. The concepts of demand latency, demand sensing, demand shaping, demand translation, and demand
orchestration are not widely understood. As a result, they are not included in the definition of corporate strategy.
Focus: Inside out, not outside in. Process focus is (today) from the inside of the organization out, as opposed to from the outside (market
driven) back. In demand-driven processes, the design of the processes if from the market back, based on sensing and shaping demand.
Vertical rewards versus horizontal processes. In supply-based organizations, the supply chain is incented based on cost reduction, procurement is
incented based on the lowest purchased cost, distribution/logistics is rewarded fro on-time shipments with the lowest costs, sales is rewarded for sell-in volume into the channel, and marketing is rewarded for market share. These incentives cannot be aligned to maximize true value.
Focus on transactions not relationships. Today, the connecting processes of the enterprise – selling and purchasing – are focused on
transactional efficiency. As a result, the greater value that can happen through relationships – acceleration of time to market through innovation, breakthrough thinking in sustainability, and sharing of demand data – never materializes.
Source: Charles Chase, SAS
®
Copyright: Jaguar-APS 52®
Copyright: Jaguar-APS
Recommendations Understand Demand
By better understanding demand, companies can plan production capacity and inventory level in a more accurate fashion, minimizing the risk of lost sales opportunities.
Collaboration and Integration The ability to share information between departments within
the business is essential to improving supply chain. Without internal communication processes in place (Demand
Planning, S&OP), the company as a whole cannot effectively collaborate with the outside entities, whether they are supplier or customers.
Supply Chain Management Increased visibility into supply decisions and constraints by
providing input in the demand shaping and shifting activity will help ensure the product is available at the right place at the right time.
Copyright: Jaguar-APS
Thank You
Charles Novak & Pavel Černý[email protected]@jaguar-aps.eu www.jaguar-aps.com
Our Solutions• Training Programs• Opportunity Assessment• Business Transformation – S&OP/IBP• Forecast Software Selection and Implementation• On-Demand
http://www.jaguar-aps.com/services/index.html