OPIM 294
Flow of material or work (Shape in Diagram)
Arrow to right
Purchase cost per unit (Symbol)
C
Component 1 of Lean Operations or Toyota System (ACRONYM)
CLOSED MITT Complexity, Labor, Overproduction, Space, Energy, Defects Materials, Inventory, Time, Transportation
Inventory Turnover formula
COGS/Inventory
Competitive Dimensions
Cost, Quality (Product and Process), Delivery Speed, Delivery Reliability, Variety, Innovation, Demand Management
Competitive Priorities
Cost, Quality - 2 most important + Responsiveness, flexibility
Disadvantages of Increasing Inventory
Costly, Consumes spaces, slows down process
Calculating Capability Index
Cpk = min {[(X-LTL)/kσ],[(UTL-X)/kσ]} Cpk< 1 process is not capable at the kσlevel Cpk≥ 1 process is capable at the kσlevel
Order Winners
Criterion that differentiates one firm from another. Examples: Cost, Service Quality, Flexibility
Order Qualifier
Criterion that permits the firm's products to even be considered for purchase. Examples: Basic quality necessary to be considered a good car (consumer reports)
Benihana is make to order
Customers come in and only then can you order, but the back kitchen could be viewed as make to stock
Cycle Time formula for processing a fixed amount of work for a specific task
Cycle Time = (Set-up Time + (Batch Size)x(Time per Unit))/Batch Size
Increasing the capacity of a non-bottleneck step can help increase the capacity of the entire system.
False
Inventory (goal definition)
all the money that the system has invested in purchasing things which it intends it sells
operational expense (goal definition)
all the money the systems spends in order to turn inventory into throughput
2 Aspects of Quality
features: more features that meet customer needs = higher quality freedom from trouble: fewer defects = higher quality
Product Development
the set of activities beginning with the perception of a market opportunity and ending in the production, sale, and delivery of the product
If the utilization approaches 0...
the waiting time approaches 0.
If the utilization approaches 1...
the waiting time approaches infinity.
Lead Time
time elapsed between placing and receiving a replenishment order
Number of Servers
n
Under Just-in-time, the ideal batch size is...
one (or low as possible)
SPC with P Charts (formulas)
p=(total number of "defects")/(number of samples, t)x(sample size, n) sp = SQRT[(p(1-p))/n] UCL = p + zsp LCL = p - zsp
Equilibrium
rate at which objects arrive = rate at which objects leave
Throughput (goal definition)
rate at which the system makes money through sales
Inventory
raw materials, component parts, work-in-process, or finished products that are held at a location in the supply chain
With Limited WIP, Production variability...
reduces the effective throughput (capacity) of the system (st. dev or variance of throughput is decreased)
Make to Order
refers to the release of work into a system only when a customer order has been received for that unit.
Reasons to Outsource - organizationa reasons
-Focus on service -Focus on core capabilities -Transform the organization -Increase flexibility
Yield Management - What is it
"Selling the right capacity to the right customer at the right price" Key: segmentation-forecasting-allocation
Effect of fast server
(1) A fast server (short processing time) has a shorter queue behind it.
Forecasting Demand: Data Requirements
(1) High-level of detail -e.g. airlines, demand by -Origin-destination pair -Fare-class -Day-of-week -Departure time (2) Quantities tracked for forecasts -Demand for each rate category/fare-class/departure -Cancellation rates -No-show rates/go-show rates -Upgrade probabilities (3) Frequent processing and updates -Daily -Even more!
Reasons to Outsource
(1) Organizational reasons (2) Operational reasons (3)Financial reasons (4)Cost driven reasons (5)Revenue driven reasons -Expand and grow with the help of another organization. -Obtain access to outsourcing partner's network
Maximizing nightly profit
= (Price per dinner - Variable costs per dinner) * Throughput - Fixed Costs
ROQ=Reorder Quantity
=EOQ (How Much?)
Factors that Complicate the forecasting Problem
(1) Seasonality -Day of week -Day of month -Time of year (2)Trends -Long-term growth -Business cycle (3) Demand censoring -Data: reservation accepted -Actual demand: includes reservations lost, spillovers (4) Special events -Sale -Super bowl
Effect of Increasing Utilization
(2) For a given level of variability, increasing utilization causes waiting time to increase at an accelerating rate.
Inventory
(L): Average number of customers (or jobs or parts) in total in the system (including queue + server) = Lq+ ρ
Inventory in Queue
(Lq): Average number of customers (or jobs or parts) waiting in the queue
Throughput Time or Lead Time - Average Total Waiting Time
(W): Average total time spent in the system by a customer (including service + waiting) = Wq+ ms
Waiting Time
(Wq): Average time spent waiting in queue
Cycle time
(ma): Average time between departures = Average time between arrivals or "inter-arrival time" = 1/λ
Average service time or processing time
(ms): 1/service rate = 1/μ
Utilization formula
(ρ): The percentage of the time the server is busy = λ/nμ where n is number of servers
Option A - Add a new floor
+ 45 Beds, -$2 million
Pull System (Pros/Cons)
-Controls maximum WIP and eliminates WIP accumulating at bottlenecks -Keeps materials busy, not operators. Operators work only when there is a signal to produce. -Throughput time and WIP are decreased, faster reaction to defects and less opportunity to create defects. -If a problem arises, there is no slack in the system. -Works better in stable environments
Reasons to Outsource - Revenue driven reasons
-Expand and grow with the help of another organization. -Obtain access to outsourcing partner's network
Push System (Pros and Cons)
-Focuses on keeping individual operators and workstations busy rather than efficient use of materials -Volumes of defective work may be produced -Throughput time will increase as work-in-process increases (Little's Law) -Line bottlenecks and inventories of unfinished products will occur -Hard to respond to special orders and order changes due to long throughput time
Reasons to Outsource - Operational Reasons
-Improve performance (quality, productivity, etc.) -Obtain expertise, skill, and technology -Risk management
Yield Management - Business Requirements
-Limited Fixed Capacity -Business environment where YM can help •Ability to segment markets •Perishable inventory •Advance sales •Fluctuating demand •Accurate, detailed information systems
How there are conflicting objectives within a single firm?
-Marketing/Sales wants: more FGI inventory, fast delivery, many package types, special wishes/promotions -Production wants: bigger batch size, depots at factory, latest ship date, decrease changeovers, stable production plan -Distribution wants: full truckload, low depot costs, low distribution costs, small # of SKUs, stable distribution plan
Reasons to Outsource - Financial Reasons
-Transfer assets to the outsourcing partner. -Free up resources for investment in other purposes.
Reasons to Outsource - Cost Driven Reasons
-Transform fixed costs in variable costs. -Reduce costs through outsourcing partner efficiencies.
General Variability as described by coefficient of variability for arrivals (ca) or coefficient of variability in service (cs)
0<ca<1 or ca> 0<cs<1 or cs>
Average number of customers in the server
1 x percentage of the time the server is busy + 0 X percentage of the time the server is idle = 1 x ρ+ 0 x (1-ρ) = ρ
IDEO - 3 Decisions for Handspring Project
1) Accept Handspring Project 2)Ask for more time then accept 3) Decline it
Decisions For Wriston Corp
1) Close the Detroit Plant and Transfer Production to other plants 2) Replace existing detroit plant with a new one 3) help improve the detroit plant 4)none of the above/do nothing
Southwest Design of Union Relationships (7 steps)
1) Deliberately slow growth; permits 2)No layoffs; allows 3) Better union relationships; creates 4) Benefit of flexible job design; fosters 5)Faster aircraft turnarounds 6)Greater aircraft utilization 7) Significantly higher profit
How to find ROP & Q
1) Find Q*=EOQ=SQRT(2DS/H) 2) Determine service level 3) Find safety factor (z-table or graph) - number of standard deviations 4) Find Std. deviation in LT demand: Squart root law) 5) Safety Stock formula: Safety factor * Std. Dev in LT Demand 6) ROP = Expected LT Demand + SS
2 Main Points about EOQ Model Uncertain Demand + Non-immediate replenishment
1) Get units shipped at a point where we account for uncertainty 2) Guarantee a service level to our customers
Key Takeaways from Wriston
1) High Overhead may be natural 2)New products first launched at Detroit Plant - low volumes, the process is not perfected yet (not standardized) 3)Prototypes are tested at Detroit plant 4)Old products/spares are made in Detroit as well(low volumes) - whenever a product's volume increases it is transferred elsewhere 5) Key takeaway: product process spectrum 6) Trade-off: volume vs. variety 7) Detroit was a job shop in the introduction stage of product life cycle - 120 distinct products made 8) We should not evaluation Wriston/Detroit based on cost - batch shop will never be as efficient - Detroit should be evaluated individually
Efficient Supply Chain Performance Measures:
1) Inventory Turnover 2) Weeks of Supply
Make to Stock - Steps to analyze the process
1) Look at the process step by step 2) Determine the processing rate (or capacity) of each step 3) Identify the process bottleneck (smallest processing rate, or largest cycle time) 4) Throughput = capacity of the bottlekneck
Make to Order - Steps to analyze the process
1) Look at the process step by step 2) Determine the processing rate (or capacity) of each step 3) Identify the process bottleneck (smallest processing rate, or largest cycle time) 4a)Determine the demand rate for the process 4b)If Demand Rate > Capacity of the Bottleneck, then Throughput = Capacity of the bottleneck, cycle time = cycle time at the bottleneck (ignore demand rate) 4c) If demand Rate < Capacity of bottleneck, then throughput = demand rate, cycle time = inverse of demand rate=1/demand rate
Mark Down Cost (3)
1) Margin/Revenue down 2) Holding costs 3) Strategic Consumer - seeking not full-price items
2 Major Herbies discovered in Alex's plant and how were they found
1) NCX-10 Machine 2) heat treat furnances They eventually identified the bottlenecks by looking for huge piles of backlogged inventory waiting in front of an operation to be processed. They tried to identify them from scheduling records, routing sheets, and other paperwork but without much luck
Stock Out Costs (3)
1) Opportunity Cost 2) Backlog Cost 3) Reputation Cost
Types of Dimensions (2)
1) Order Qualifier 2) Order Winners
Attacking and Defending through Operations (2)
1) Positioning 2) Capabilities
Characteristics of Successful Product Development (5)
1) Product Quality 2) Product Cost 3)Development Time 4)Development Cost 5)Development Capability
3 Ways to Match Supply with Demand
1) Redesign the supply chain 2) manage the demand 3) coordinate the supply chain
Responsive Supply Chains do 3 things:
1) Reduce uncertainty 2) Avoid Uncertainty 3) Hedge against uncertainty
Physically Efficient Supply Chain - 1)Primary Purpose 2)Inventory Strategy 3)Lead-Time Focus, 4) Choosing Suppliers 5) Product-Design
1) Supply predictable demand efficiently at lowest possible cost 2) Generate high turns and minimize inventory, 3) Shorter only if it doesn't raise costs 4)Cost & Quality 5) Max. performance at min. cost
When to use P-charts
1) When decisions are simple "yes" or "no" by inspection 2) when the sample sizes are large enough (>50)
IDEO decline Handspring Project
1) conflict of interest - Palm is a client 2) Covertness - disturbs/contaminates culture 3) hurt's IDEO's reputation of being an innovator 4) breakthrough innovations require more time
IDEO ask for more time then accept Handspring Project
1) don't have to sacrifice legendary process (breakthrough innovations require more time) 2)better product with more time since don't have to sacrifice style and and settle on an inexpensive plastic housing/AAA batteries
when to use make to order system
1) products or parts are processed in low volume and high variety 2)customers are willing to wait for their order 3)it is expensive or difficult to store the flow units
Issues (3)
1)Market 2)Technical 3)Resource
Variations to the Product Development Process (6)
1)Market Pull Products 2)Technology Push Products 3)Platform Products 4)Process Intensive Products 5)Quick build Products 6) Complex Systems
Market Responsive Supply Chain - 1)Primary Purpose 2)Inventory Strategy 3)Lead-Time Focus, 4) Choosing Suppliers 5) Product-Design
1)Respondquickly to unpredictable demand to minimize stockouts, markdowns. 2) Deploy significant buffer stocks of parts or FGI 3) Aggressively try to reduce lead time 4) Speed, Flexibility, Quality 5) Modular, try to delay differentiation
Challenges of Product Development (6)
1)Trade-offs 2)Dynamics 3)Details 4)Time Pressure 5)Economics 6)Organizational realities
Supply chains perform two roles:
1. physical conversion of raw materials into goods (functional products) -production, transportation and inventory costs 2. market mediation (matching supply and demand) (Innovative) -markdown losses or shortage costs
Product Development Process (also called new product development funnel)
1a) Customer needs and Technological Possibilities; leads to 1b)Product design 2)Prototyping & testing 3a)Pilot Production 3b)Manufacturing ramp-up and release 4) Launch
Challenge 2: Design the bar
79 seats was optimal - the bar adds inventory of customers
If the setup time of a task is extremely large, would you prefer a small or a large batch size?
A large batch size to split the setup time over.
Sigma
A measure of Variability
P Control Charts
A plot of proportions over time (used for performance measures that are yes/no attributes) They help to calculate acceptable limits and help with statistical process control
Transformation/Production Process
A transformation/production process uses resources to convert inputs into some desired output
Option B - Add a Saturday shift
Against the culture - shouldice staff wouldn't like it
Adv. and Dis. of Outsourcing
ADv. cheap labor, focus on core competency, save fixed costs, bypass startup ideas disadv.: transaction cost, intellectual property, quality + design, more variable costs
How Benihana works to address high variability of customer demand
Add Inventory with the bar, Batching and the happy hours reduces variability, reducing the dining time reduces utilization
Challenge 6: Best combination
All 5 things - each is trying to reduce variability but in different ways - whether to add inventory, reduce variability, or reduce utilization.
Utilization
Average proportion of time the step is working on something = capacity of process (bottleneck throughput rate)/capacity of task
Cycle Time
Average time for completion of a unit at a production step. Measured as time/unit. Cycle Time = (Set-up Time + (Batch Size) x (Time per unit))/Batch Size
Little's Law
Avg Throughput time = Avg WIP/Avg Processing Rate W=L/λ or L =λW Applies to Bottlekneck
Shouldice bottleneck
BEDS - but could shift if they increase the capacity of the beds
Capacity Given Batch Size (formula)
Batch Size/(Setup Time+(Batch Size)x(Time Per Unit))
Capabilities (3)
Being better at "the same game" - (1) Process Based (2) System (Coordination) based (3) Organization based
Shouldice process
Blood checks -> exams -> Accounting/Processing -> Beds -> operating room -> surgeons
Which production step limits the process capacity?
Bottleneck
Bottlekneck
Bottleneck is the process stage with the lowest processing rate (highest cycle time)
How can we monitor quality?
By observing variation in output measures! 1.Assignable variation: we can assess the cause 2.Common variation: variation that may not be possible to correct (random variation, random noise)
How many units per unit time can go through each task? or through the process as a whole?
Capacity (or processing rate)
Performance Measures of Efficiency (4)
Capacity (or throughput rate), Throughput (or throughput rate), bottleneck, throughput time
Calculation for Net Benefit
Change in Waiting Time = W(old) - W(new) Net Benefit = Change in W x Savings/Time/Unit x λ
Walmart Distribution Decisions
Close location of stores to warehouses Hub and spoke model Logistics optimization
Better Measure of Variability
Coefficient of Variation c=σ/mean(m) Proportional to number of people in the place
Cost of Quality
Cost of quality can be 10-15 % of sales and include 1.Prevention costs 2.Appraisal costs 3.Internal failure costs 4.External failure costs 5.Opportunity costs
Horizontal Integration
Consolidation of power within industry Competition is based on "monopolization" of the industry Antitrust Issues might arise Can deter entry Ex. Financial Services, Hi-Tech industries, airlines Adv: economies of scale, eliminate pricing war Disadv=: antitrust, costly
Why Should firms not outsource?
Coordination reasons -Messy interfaces -Highly specialized information or technology Strategic Control reasons -Control over specific assets -Brand impacts -Long term R&D Intellectual Property reasons -Weak IP protection -Technology is easy-to-imitate
EOQ Model - Number of Orders=Order Frequency
D/Q
Ordering Cost (formula)
D/QxS number of orders x cost per order
6 Sigma Roadmap
DMAIC Define, Measure, Analyze (and fix the obvious), Improve, Control (move to next project)
Alternative paths
Decision - you only go through one path so you are focused on through. Grocery Store. Assume same rate for 2 pathways then btw. A&B = rate/2 Use the cycle time for a single path that produces one unit of the product at a time
Trade-offs
Decisions that arise because of the inability of processes to excel simultaneously across all competitive dimensions
Which Supply Chain to Use?
Depends if functional (Campbell Soup) or Innovative (Sport Obermeyer, Zara) products
Decision Point
Diamond
Exponential Distribution
Distribution of inter-arrival and service times where coefficient of variation is equal to 1 - standard deviation (σ) is equal to mean in exponential distribution
Benihana Bottleneck
During peak is capacity (# of seats) during pre/post-peak it is demand.
Purchasing Cost (Formula)
DxC Total Units x cost per unit
Southwest Culture - Evidence
Egalitarian Approach - To customers and employees Fun - Senior Leaders example, Employee Selection (Most embarassing moment, stand during interview, have customers observe interviews) Teamwork - Group incentives/shared goal/flexible scheduling
Efficient Supply of Functional Product: Campbell Soup
Electronic Data Interchange with retailers to ensure continuous replenishment and lower inventories.
Flow Shop
Equipment in process sequence. Expensive equipment but high utilization rates. (e.g. refinery, steel mill). Mass Production is the production of large amounts of standardized products, including and especially on assembly lines. With job production and batch production it is one of the three main production methods
Assembly Line
Equipment placed in assembly sequence. (e.g. automobile manufacturer). A manufacturing process (most of the time called a progressive assembly) in which parts (usually interchangeable parts) are added as the semi-finished assembly moves from work station to work station where the parts are added in sequence until the final assembly is produced.
Push System (How it works)
Every worker maximizes own output, making as many products as possible
Option C - Build a new facility
Expensive, hard to replicated process, institutional knowledge, but could possible to move closer to customers in the US, also there could be different government regulations
An hour lost on a bottleneck is an hour lost on the system
False
Throughput of a process > Bottleneck capacity
False
Challenge 4: Boost Demand w/Advertising and Special Programs
Happy hour - demand smoother of when and how people arrive, other 2 programs only boost demand but are too expensive. The peak time is the most profitable, but there this too much demand
Queue+server
Flow rate = λ Wait = W # = L Lq= λ W
Queue Only
Flow rate = λ Wait = Wq # = Lq Lq= λ Wq
Customer Flow at Benihana
Flow shops - Maximize efficiency, compared to job shop which maximizes flexibility
Jeff Hawkins
Founder of Palm Computing - sought out IDEO for Palm
David Kelley
Founder/CEO of IDEO
Push System (Diagram)
From RM to Final Assembly/FGI/Customer, at each step, information flow and material flow go from RM to Final Assembly with no information flowing back to RM.
Pull System (Diagram)
From RM to Final Assembly/FGI/Customer, at each step, information flows backward from Final Assembly to RM while material flows forward from RM towards FGI
Batch Shop
Groups similar equipment together. Batches provide some standardization(e.g. copy center making 10,000 copies of an ad piece for a business)
Job Shop
Groups similar equipment together. Jobs follow diverse paths. (e.g. copy center making single copies of student term papers). Typically small manufacturing systems that handle job production, that is, custom/bespoke or semi-custom/bespoke manufacturing processes such as small to medium-size customer orders or batch jobs. Job shops typically move on to different jobs (possibly with different customers) when each job is completed.
Littlefield 1: Product Lifecycle for DSS
Growth, Maturity, Decline "In the initial months, demand is expected to grow at a roughly linear rate, stabilizing after about 5 months. After another month, demand should begin to decline at a roughly linear rate for 3 months."
Annual holding cost of average inventory (symbol)
H
Walmart Merchandising/Marketing/Stores/Pricing Decisions
High Variety Price-matching strategy -EDLP (Everyday Low Prices) - (marketing maneuver + reducing variability) Mostly hard products (not many perishable goods) -more $ per space National brands and private label (called Great Value)
Advantage of Job Shop
High flexibility in product engineering High expansion flexibility (machines are easily added or substituted) High production volume elasticity (due to small increments to productive capacity) Low obsolescence (machines are typically multipurpose, not specialized/expensive) High robustness to machine failures
Why Shouldice was Successful
High quality/low recurrence rate, Cheaper, High customer satisfaction, low turnover of staff/doctors, quicker recovery with the Shouldice method, bonus linked to performance, highly profitable
Effect of Increasing Initial Inventory from 4 to 8 (Scenario 2 in poker chips)
Higher average Throughput, higher standard deviation of throughput
Push System - Throughput Time, WIP Inventory, Ability to Responds to customized or special orders
Higher, Higher, Lower
From The Goal: Alex tries to explain to Hilton that an hour lost on a bottleneck is an hour lost on the system. Why does Hilton disagree? (no more than 4 sentences please!)
Hilton is operating under the traditional assumption that every piece of equipment or operation should be evaluated on its stand-alone efficiencies, costs, and production output. Those assumptions ignore the two phenomena of dependent events and statistical fluctuations that link the equipment together into a production system. Because of these phenomena, disruptions on one piece of equipment affects the whole system.
Make to order process
If the system is make to order, throughput = MIN(Demand, Capacity)
Calculating the Coefficient of Variation
Inter-arrival times -mean = ma= 1 / λ -standard deviation = sa -ca= (std. dev.) / mean = sa/ ma Service times -mean = ms= 1 / μ -standard deviation = ss -cs= (std. dev.) / mean = ss/ ms
Problem with EOQ Model
It is highly insensitive to its parameters such as human error Ex. double holding costs Optimal per unit total only moves slightly because of square root . Remember slides
Capability Index ( Cpk)
It shows how well the performance measure fits the design specification based on a given tolerance level
Buffering
Keep some inventory between stages. ex. 1 unit in buffer of capacity 2.
average # of items in the system
L
Dennis Boyle
Leader of Palm V Project at IDEO
How Long does it take you to transform a dollar invested in inventory into sales? (Hopefully profitable)
Little's Law: L=λW Inventory=Flow Rate(COGS) x Flow Time (Inventory/COGS=1/Inventory turnover)
Functional - Product Life Cycle, Contribution Margin((Price-VC)/Price), Averagemargin of error in the forecast at the time production is committed, Average Stock out rate, Average forcedend of season markdown as percentage of full price
Long, Low, Low, Low, Low
Southwest Emphasis (10)
Low Price Fast Turnaround Simple Fares P2P (instead of hubs) Frequent short-haul flights Friendly interactions convenient airports rewards program multi-tasking employees uniform process and plane type (Boeing 747)
How are Benihana's costs different?
Low Variety in menu selection, Labor - not many people in the back room, not many waitresses
Smaller Batches translate to...
Lower Inventory Levels, lower wait time, losses in capacity
Effect of Just Reducing Variability (3 and 4 are 50% probabilities for each machine rather than 1-6 possibilities)
Lower Standard Deviation, higher average throughput rate than just increasing initial inventory
Effect of increasing initial inventory by 8 (from 4) Reducing Variability (3 and 4 are 50% probabilities for each machine)
Lower Standard deviation (same as scenario 3), higher average throughput rate
Pull System - Throughput Time, WIP, Ability to Responds to customized or special orders
Lower, Lower, Higher
Lq (formula)
Lq = ρ^(SQRT(2(n+1))/(1-p)*(((ca^2)+(cs^2))/2)
Goal
Make Money
High Time Pressure, Low Fixed Cost of Prototyping
Mixed Strategy
Low Time Pressure, High Fixed Cost of Prototyping
Mixed Strategy
Outsourcing
Moving some of the firms internal activities and decisions to outside providers (domestic or offshore)
Characteristics of Assembly Line/Flow Shop
Narrow Product Variety, Low Product Quality, Less Time spent on Product, Low Price High Volume per product, High structure on equipment and labor ex. McDonalds
Expected Payoff Newsvendor
Newsvendor will purchase the copy at X if it thinks underage cost is higher than overage cost. Pr(Sold) Cu - Pr(not sold)co >=0 Let Pr(not sold) = p, then Pr(Sold) = 1-p Newsvendor should purchase the Xth copy if and only the expected payoff is positive p<= cu/(cu+co)
Is the queue/buffer alwaysempty?
No
95% Service Level With Uncertain Demand + Non-immediate replenishment
Normal(D xL, σL2). 5% chance of stock out SS = z σL ROP = D xL + SS = D xL + z σL. Same EOQ.
IDEO Process
Phase 0 -Understand Phase Phase 1 - Visualize and Realize Phase 2 - Evaluation and Refinement (Human factors to engineering Phase 3 - Implementation (detailed engineering started) Phase 4 - Implementation (manufacturing liaison) -
Time between success orders
Number of Weeks in Year = 52 52/(D/Q)
Throughput Rate
Number of customers served/unit time = λ= arrival rate = departure rate
Capacity
Number of customers that can be served/unit time = μ
Booking Limit Model
Objective: Solve for the booking limitwhich is the number of rooms you are willing to sell at the discount rate. A single decision is made before uncertain demand is realized Similar to Newsvendor Model
Challenge 5: Use different types of batching
Optimal: 8, 8, 4 to 8
High Time Pressure, High Fixed Cost of Prototyping
Parallel
High Volume, High Standardization
Process = Flow Shop Point in Product Life Cycle = Entry
High Customization, Low Volume
Process = Job Shop Point in Product Life Cycle = Entry
Computing Cycle Times
Processing a fixed amount of workfor a specific task.
Phase 1 IDEO
Product direction is chosen through visualized potential solutions through tangible prototypes; 3D models made, understanding of context in which product is used, and outline of manufacturing strategy; storyboards instead of stats through market research
Pull System (How it works)
Production line is controlled by the last operation, Kanbancards control WIP
Partial Productivity Measures
Productivity = Output/Labor=Output/Capital=Output/Material
Productivity (formula)
Productivity = Outputs/Inputs(resources)
IDEO Accept Handspring Project Pros and Trade-offs
Pros 1) Leverage Palm V Experience 2) Simpler Project - Palm operating system licensed to Handspring, Low operating costs, limited/reduced time, follow client's orders, triple AAA bateries instead of lithium-ion batteries found in Palm V, challenge to firm 3) Customer is king - don't want to lose a client Tradeoffs If IDEO fails Handspring from accepting immediately, they can learn how to innovate on incremental products They could also lose future clients from not knowing how to innovate on incremental products
Total Cost (Formula with Words)
Purchasing Cost + Ordering Cost + Inventory Cost
Decision Variable in EOQ Model
Q* We use the optimal quantity to optimize the total cost with respect to the decision variable (the variable we control)
Average Inventory for each period is
Q/2
Inventory Cost (formula)
Q/2xH average inventory x holding cost
Value
Quality/Cost
In The Goal, what are the three primary financial measurements used to know whether the company is making money?
ROI, Net Profit, and Cash Flow
Deterministic Demand + non-immediate replenishment - EOQ model When to order?
ROP = LxD L=Lead time in periods=Receive Order - Place Order D=Demand per period Deterministic Demand + Lead Time (L) >= 0 If demand is known exactly, place an order when inventory equals demand during lead time
Processing Step ( Shape in Diagram)
Rectangle
Everyday Low Prices (EDLP)
Reduce variability marketing "bull-whip effect" If give periodic discount, then retailers will buy and stock up.
Effect of Reducing Variability
Reduces the standard deviation of the throughput
R&D Productivity (formula)
Revenue/R&D Expenses
SGA Productivity (formula)
Revenue/SGA
Overall Productivity (formula)
Revenue/Total Expenses
RRR
Rough - cheap Rapid - fail often Right - getting specific needs of customer
Order Cost (Symbol)
S
Q*
SQRT(2DS/H)
What Southwest Gives Up (7)
Seat Assignments Locations Meals Baggage Transfer First/Business Class Upgrades Variety
Low Time Pressure, Low Fixed Cost of Prototyping
Sequential
Deterministic
Service Time or Arrivals are held constant. No Randomness - complete certainty about the value of the random variable. Coef. of variation for arrivals = (std. dev.) / mean = ca= 0 Coef. of variation in service = (std. dev.) / mean = cs= 0
Innovative - Product Life Cycle, Contribution Margin((Price-VC)/Price), Averagemargin of error in the forecast at the time production is committed, Average Stock out rate, Average forcedend of season markdown as percentage of full price
Short, High, High, High, High
Challenge 3: Change Dining Time
Smooths out demand
Vertical Integration
Supplier's abilities are key to competitiveness Competition requires capital intensive assets Collaboration problems and supply risks are reduced. Double Marginalization is removed Adv: Reduce transaction cost, reliable supply Disadv: costly
Starving
Stoppage of activity because of lack of material. empty buffer.
Blocking
Stoppage of flow because there is no storage place. full buffer.
How did Shouldice become so successful?
System geared toward efficiency, flow shop/assembly line - correct process structure, they pre-screen patients with one type of survey, appointment only, only one type of patient - external abdominal hernias - selection bias, train surgeons to use exactly the same procedure. All this helps to reduce variability, making costs go down and throughput go up.
Total Cost (Formula w/ Symbol)
TC = D*C + (D/Q)*S+(Q/2)*H
Throughput Time
The Average time it takes for a unit to make it through a task/system. The time it take for a unit to go from start to finish of processing. Sum of Processing Steps (Add to Rectangles in Process Flow)
Challenge 1: Batch?
The Capacity remains the same, but there is a higher throughput = higher utilization
Capacity of the Process
The capacity of the process is: •minimum processing rate across all stages •capacity of the stage with the longest cycle time
What is a supply chain?
The concept of supply chain defines a series of companies interlinked together by a flow of material and information in order to serve the end customer. Contrast with a process which is just 1 company
Hypbrid/Mixed Prototype
The design team builds and tests multiple prototypes each period, for as many periods as are necessary to achieve a "good enough" design.
Parallel Prototype
The design team builds and tests multiple prototypes in a single period, then chooses the most profitable one.
Sequential
The design team builds and tests one prototype each period, for as many periods as are necessary to achieve a "good enough" design.
Make to Stock
The firm finds itself in this situation because it commits to its entire supply before demand occurs. This mode of operation is often called make-to-stock because all items enter finished goods inventory (stock) before they are demanded. In other words, with make-tostock, the identity of an item's eventual owner is not known when production of the item is initiated
Capacity or Processing Rate (of a step)
The maximum flow rate wat which a process can operate. Average number of units processed over a time interval. Measured as units/time. Processing Rate = 1/Cycle Time
Service Level (definition of EOQ calc. Uncertain Demand + Non-immediate replenishment
The probability of not stocking out
Productivity Measurement
The productivity index is a relative measure. It has to be compared with something else: 1) benchmarking - competitors 2) changes across time 3) normalized date
How many units per unit time does the process produce?
Throughput (or throughput rate)
How long does it take for a unit to get through the system?
Throughput Time
Importance of Inventory
To summarize, we build and keep inventory in order to match supply and demand in the most cost effective way.
Shouldice Problem
Too much demand - afraid to do marketing, backlog of patients already waiting - they needed solutions/options to expand capacity
How JIT accounts for problems of too much inventory stored
Too much inventory hides problems... Problems with inventory include: WIP queues (banks), paperwork backlog, scrap, engineering design redundancies, inspection backlogs, machine downtime, vendor delinquencies, change orders, design backlogs, decision backlogs
Quality Improvement
Traditional - Cyclical - Quality decreases then gets better Continuos Improvement - steady improvement with no cyclicality and major dips in quality
RM Process: How to allocate capacity?
Typical capacity control mechanisms -Ad hoc negotiation, booking limits, "bid prices" or hurdle rates Ultimately, an accept-deny decision for each service request How to properly quantify the decision?
Phase 0 IDEO
Understand everything about new client and its business; look at history of product and test existing products such as every remote that is on the market; look at consumer observations like people at home on their couches attempting to use their remotes; by end, team created a feasibility record along with major discoveries about the marketplace and users
Holding - RM, WIP, FGI (Shape in Diagram)
Upside down triangle
Std. Deviation in LT Demand
Variance over multiple periods = the sum of the variances of each period (Assuming independence) Standard deviation over multiple periods is the square root of the sum of the variances not the sum of the standard deviations
2 Ways to Design Supply Chain
Vertical Integration (Ex. Automotives, Aerospace) Horizontal Integration (Ex. Financial Services, Hi-tech Industries)
Walmart Procurement Decisions
Very cost-focused: -minimize expenses -intense negotiations -bypass wholesalers -buy low & stock up -global procurement Online info exchange in real-time Scan N' Pay (also good for suppliers)
Disadvantage of Job Shop
Very hard scheduling due to high product variability and twisted production flow Low capacity utilization
average time an object spends in the system.
W
Walmart Takeways
WMT makes 3 types of decisions to make their supply chain as efficient as possibly: Procurement, Distribution, Mechandising/Marketing/Stores/Pricing WMT sells functional products so they are focused on beining efficient as possible by driving down any cost in their supply chain
Measuring Variability
We know standard deviation ("σ") is a measure of variability
EOQ - Uncertain Demand + Non-immediate replenishment Distribution of Demand during 1) over some period 2) lead time
When the given demand (over some period) is random: Normal(D, L*σD^2), demand during lead time is also random: Normal(Lx D, L*σD^2). In other words,demand during lead time has a normal distribution with mean Dx L and standarddeviation σL=SQRT(L*σD^2)
Why not choose the largest possible batch size to maximize capacity?
While large batch sizes are desirable from a capacity perspective, they typically require a higher level of inventory, either within the process or at the finished goods level. Holding the flow rate constant, we can infer from Little's Law that such a higher inventory level also will lead to longer flow times (longer wait times - slows down process)
Characteristics of Job/Batch Shop
Wide Product Variety, High Product Quality, More Time spent on Product, High Price Low Volume per product, less structure on equipment and labor ex. Tombs
When a process is kσ capabale
X+kσ<=UTL 1<=(UTL-X)/kσ X-kσ<=LTL (X-LTL)/kσ>=1
Why can't achieve goal by focusing on efficiency
You can make system more efficiency by increasing the efficiency of a non-bottleneck operation (e.g. robots) but that does not help the firm make more money and can sometimes work against the goal. Efficiency does not necessarily increase throughput, reduce operating expenses, or reduce inventory
Organization Based
ability to master new technology, design, and bring new plants faster
System (Coordination) Based
better at managing supply chain partners
Process Based
better at transforming material or information
Coefficient of variability for arrivals - symbol
ca
Overage Cost
co= overage cost = loss margin -The cost of ordering one more unit than what you would have ordered had you known demand -In other words, suppose you have left over inventory (i.e., you over ordered). co is the increase in profit you would have enjoyed had you ordered one fewer unit. -For our newsvendor, co= Cost -Salvage value = $.7-$.2=$.5
Simultaneous
coffee shop. Between A&B = max (upper,lower) Both stages have to be finished before the cycle is complete so take the max time/unit of upper and lower branches.
Underage Cost
cu= underage cost = profit margin -The cost of ordering one fewer unit than what you would have ordered had you known demand -In other words, suppose you had lost sales (i.e., you under ordered). cu is the increase in profit you would have enjoyed had you ordered one more unit. -For our newsvendor, cu= Price -Cost = $1-$.7=$.3
In The Goal, the theory of bottlenecks re-prioritizes the traditional importance of improvement opportunities to:
d. 1. Throughput 2. Inventory 3. Operating Expense
Phase 2 IDEO
develop functional prototypes; resolve technical problems and problems faced by users; emphasis shifts from human factors to engineering and ergonomics; at end, functional model as well as a "looks like" design model is delivered; industrial design solutions becomes documented using CAD tools Technical specifications need to be finalized to do detailed engineering which occurs in Phase 3
Efficiency
doing something at the lowest possible cost
Effectiveness
doing the right things to improve the quality of the products/services
IDEO Process
first-hand knowledge, deep dive, wild ideas first, command decision, adults refocus, no time constraints, prototype driven, RRR (rough rapid right), enlighted trial and error, when meeting client for first time - don't show them everything
IDEO Culture
flat - lack of hierarchy, don't hire people that don't listen, ask for permission, diversity, team, democratic process, playful environment, experimentation, peer chosen reviews, understand/observe
The higher the variability...
higher the delay - Variability causes delays.
Innovative products exhibit:
highly uncertain market behavior
Weeks of Supply
how many weeks worth of inventory does the company have on hand. High value of weeks of supply means that the firm has a lot of inventory sitting around.
Inventory Turnover
how often the company replenishes inventory. High value of inventory turnover means that the inventory was not sitting around a long time
Increasing the batch size...
increases the capacity
ROP=Reorder Point
inventory balance at which your order is triggered (When?)
Job Production
involves producing custom work, such as a one-off product for a specific customer or a small batch of work in quantities usually less than those of mass-market products
If the setup occurs at a nonbottleneck step (or the process is demand-constrained)
it is desirable to decrease the batch size, as this decreases inventory as well as flow time.
If the setup occurs at the bottleneck step (and the process is capacity-constrained),
it is desirable to increase the batch size, as this results in a larger process capacity and, therefore, a higher flow rate.
Laying off people doesn't necessarily move the firm toward its goal because
it reduces operating expense but doesn't increase throughput or reduce inventory
Innovative products
short lifecycle products, products that require continuous innovation, unpredictable demand, high margins. E.g. Sport Obermeyer, Zara, Fashion Goods, consumer electronics
Functional Products
simple, commodities, stable demand patterns, long product life cycles, low product variety, low margins. E.g. Food products such as Campbell Soup
Phase 4 IDEO
smooth product release to manufacturing ensured; testing manufacutring feasibility is crucial - each day's loss of production line's output might cost the client company a substantial amount in lost revenues; by end, the product is formally handed over to client
What does Happy Hour do?
solves 2 problems - improves the situations across all times - smoothes demand
Functional products exhibit:
stability in market behavior
Phase 3 IDEO
team completes product designs and verified that the product worked; engineering predominates; design team visits machine shop;team delivers fully functional design model, tooling databases, and technical documentation; testing for govt regulations; selecting vendors
Supply Chain Management (definition)
the design and management of processes across organizational boundaries with the goal of matching supply and demand in the most cost effective way.
Service Level
the fraction of replenishment cycles over which you don't stock out.
average # of items in the system
λ
Average Arrival Rate
λ customers/unit time
Case to focus on
λ< μ
Average Service Rate
μ customers/unit time
Summary of Lean Operations
• know the Pros and cons of Lean Operations/Toyota System. •Lean Operations is the goal to achieve high quality at minimum cost and inventory. •The flow of material is pulled through the process by downstream operations. •Lean originated with the Toyota Production System: elimination of waste and just in time.
6 Sigma Quality
•A philosophy and set of methods companies use to eliminate defects in their products and processes •Seeks to reduce variation in the processes that lead to product defects •The name "six sigma" refers to the variation that exists within plus or minus six standard deviations of the process outputs 6sigma = 99.9997% X+6σ must occur before acceptable limit (see slides)
Lean Operations or Toyota System (Definition)
•A system that continually searches for and eliminates waste throughout the value chain. •Views every enterprise activity as an operation and applies its waste reduction concepts to each activity -from Customers to the Board of Directors to Support Staff to Plants to Suppliers.
Lean Operations (What it does)
•Attacks waste •Exposes problems and bottlenecks •Achieves streamlined transformation
Process Quality
•Conformance •Defects
What do you consider when deciding how much to buy?
•Cost of not having it. •Cost of going to the grocery or gas station (time, money), cost of drawing money. •Cost of holding and storing, lost interest. •Price discounts. •How much you consume. •Some safety against uncertainty.
Economic Order Quantity Model (EOQ)
•Demand,D, is known and deterministic on a yearly basis •We have a known ordering cost, S, and immediate replenishment •Annualholding costof average inventory is H •Purchasing cost C per unit •no lead time
Lean Operations (What is Requires)
•Employee participation •Continuing improvement •Quality control •Small lot sizes
What do we learn from the Newsvendor
•Forecasts are always wrong. A demand estimate that only gives the mean is too simple, you also need the standard deviation. •If cu= cothenp = cu/ (cu+co) = .5 andz = 0. We will have X = μ, i.e., one should order the expected demand. •When making a "bet", also consider the underage against the overage cost. -When underage cost is higher than overage cost, bet MORE than the expected demand (i.e., cu > co => X > μ) -When overage cost is higher than underage cost, bet LESS than the expected demand (i.e., cu < co => X < μ) •The smaller the standard deviation, the closer will be the optimal order quantity to the expected demand.
Benefits of Inventory
•Hedge against uncertain demand •Hedge against uncertain supply •Economies of Scale •Workload Smoothing
Inventory Costs from Holding
•Holding (carrying) costs ($) •Setup costs ($)
Lean Operations (What it is)
•Management philosophy •"Pull" through the system
Lessons from Littlefield 1
•Matching flow with demand is a continuous effort. •Watch bottlenecks and capacity-constrained resources closely. •To make capacity decisions, estimate the cost/benefits of -having "too much" or "too little" capacity. -being early or late in adding (or reducing) this capacity. •Queues can get huge as you approach high capacity utilization. •High utilization may not be helpful. •Remember "The Goal": reducing inventories, increasing throughput rate and reducing operating costs are keys to profitability and good cash flow. •Have a strategy (plan), be proactive (not reactive).
Pros of JIT
•Minimal inventory •Less space •More visual •Easier to spot quality issues
Newsvendor Framework
•One chance to decide on the stocking quantity for the product you're selling. •Demand for the product is uncertain. •Known profit margin for each unit sold and known loss margin for the ones that are bought and not sold. •Goal: Maximize expected profit
Component 2 of Lean Operations or Toyota System
•Only produce what's needed •The opposite of "Just In Case" philosophy •Ideal lot size is one •Minimize transit time •Frequent small deliveries
Inventory Costs on Supply Side
•Ordering costs ($) •Setup costs/time •Purchase costs ($) •Delivery times
Product Quality
•Performance •Reliability/Durability •Serviceability •Features •Aesthetics
Costs of Inventory
•Physical holding costs: -out of pocket expenses for storing inventory (insurance, security, warehouse rental, cooling) -All costs that may be entailed before you sell it (obsolescence, spoilage, rework...) •Ordering costs •Shortage costs: Lost revenues, customer goodwill. •Opportunity cost of inventory: foregone return on the funds invested. •Operational costs: -Delay in detection of quality problems. -Delay the introduction of new products. -Increase throughput times.
Cons of JIT
•Requires discipline •Requires good problem solving •Suppliers must be close •Requires high quality
Inventory Costs on Demand Side
•Shortage costs ($) •Salvage losses ($) •Demand levels
Lean Operations (What it Assumes)
•Stable Environment
Completely Random
•Time between arrivals or service times are distributed exponentially •#of arrivals or #of services in a given interval is distributed Poisson •Coef. of variation = (std. dev.) / mean = ca= 1
Why is it so Difficult to Match Supply and Demand
•Uncertainty in demand and/or supply •Changing customer requirements •Decreasing product life cycles •Fragmentation of supply chain ownership •Conflicting objectives in the supply chain •Conflicting objectives even within a single firm