1. Four Dimensions of Performance: Trade-offs
   1. Cost
      1. Efficiency
      2. Measured by:
         1. cost per unit
         2. utilization
   2. Quality
      1. Product quality (how good?) => Price
      2. Process quality (as good as promised?) => Defect rate
   3. Variety
      1. Customer heterogeneity
      2. Measured by:
         1. number of options
         2. flexibility / set-ups
         3. make-to-order
   4. Time
      1. Responsiveness to demand
      2. Measured by:
         1. customer lead time
         2. flow time
2. Process Analysis
   1. Processes: The Three Basic Measures
      1. Flow Unit: Customer or Sandwich
      2. Flow rate / throughput: number of flow units going through the process per unit of time
      3. Flow Time: time it takes a flow unit to go from the beginning to the end of the process
      4. Inventory: the number of flow units in the process at a given moment in time
      5. Example
         1. Immigration department
            1. Flow Unit: Applications
            2. Flow rate: Approved or rejected cases
            3. Flow Time: Processing time
            4. Inventory: Pending cases
         2. Champagne
            1. Flow Unit: Bottle of champagne
            2. Flow rate: Bottles sold per year
            3. Flow Time: Time in the cellar
            4. Inventory: Content of cellar
         3. MBA program
            1. Flow Unit: Student
            2. Flow rate: Graduating class
            3. Flow Time: 2 years
            4. Inventory: Total campus population
         4. Auto company
            1. Flow Unit: Car
            2. Flow rate: Sales per year
            3. Flow Time: 60 days
            4. Inventory: Inventory
   2. Finding the bottleneck
      1. Basic Process Vocabulary
         1. Processing times: how long does the worker spend on the task?
         2. Capacity=1/processing time: how many units can the worker make per unit of time If there are m workers at the activity: Capacity=m/activity time
         3. Bottleneck: process step with the lowest capacity
         4. Process capacity: capacity of the bottleneck
         5. Utilization=Flow Rate / Capacity
         6. Flow rate=Minimum{Demand rate, Process Capacity}
         7. Flow Time: The amount of time it takes a flow unit to go through the process
         8. Inventory: The number of flow units in the system
      2. Capacity Calculations
         1. Capacity_i=Number of Resources_i/Processing Time_i
         2. Process Capacity=Min{Capacity_i}
         3. Flow Rate=Min{Demand, Capacity}
         4. Utilizationi=Flow Rate/Capacity_i
      3. Labor productivity measures
         1. Cycle time CT= 1/ Flow Rate
   3. Little's Law
      1. Processes: The Three Key Metrics
      2. What it is: Inventory (I) = Flow Rate (R) * Flow Time (T)
      3. How to remember it: - units
      4. Implications:
         1. Out of the three fundamental performance measures (I,R,T), two can be chosen by management, the other is GIVEN by nature
         2. Hold throughput constant: Reducing inventory = reducing flow time
      5. Given two of the three measures, you can solve for the third: Indirect measurement of flow time: how long does it take you on average to respond to an email? You write 60 email responses per day You have 240 emails in your inbox
      6. Example
         1. In a large Philadelphia hospital, there are 10 births per day. 80% of the deliveries are easy and require mother and baby to stay for 2 days 20% of the cases are more complicated and require a 5 day stay What is the average occupancy of the department?
         2. R = 10 babies/day
         3. T = 0.8 * 2 days + 0.2 * 5 days = 2.6 days
         4. I = R * T = 10 babies/day * 2.6 days = 26 babies
      7. Some remarks
         1. Not an empirical law
         2. Robust to variation, what happens inside the black box
         3. Deals with averages – variations around these averages will exist
         4. Holds for every time window
         5. Shown by Professor Little in 1961
   4. Inventory Turns / Inventory costs
      1. Inventory Turns
         1. T = 391/20000 * 365 = 7 days
         2. T = 29 days
         3. Inventory turns = COGS/Inventory
            1. Careful to use COGS, not revenues
      2. Inventory Turns in Retailing and Its Link to Inventory Costs
         1. Per unit Inventory costs= Annual inventory costs/Inventory turns
         2. Example:
            1. Annual inventory costs=30%
            2. Inventory turns=6
            3. Per unit Inventory costs= 30% per year/6 turns per year = 5%
      3. Buffer or Suffer
         1. Simple Process Flow – A Food Truck
         2. Buffer-or-suffer strategy
            1. Buffering is easier in production settings than in services (make to order vs make to stock) Preview two different models: Queue and Newsvendor
         3. Difference Between Make-to-Order and Make-to-Stock
            1. McDonald's
            2. 1. Make a batch of sandwiches
            3. 2. Sandwiches wait for customer orders
            4. 3. Customer orders can filled immediately
            5. => Sandwich waits for customer
            6. Subway
            7. 1. Customer orders
            8. 2. Customer waits for making of sandwich
            9. 3. Customer orders can filled with delay
            10. => Customer waits for sandwich
            11. Which approach is better?
            12. Make-to-Stock advantages include:
            13. + Scale economies in production
            14. + Rapid fulfillment (short flow time for customer order)
            15. Make-to-Order advantages include:
            16. + Fresh preparation (flow time for the sandwich)
            17. + Allows for more customization (you can't hold all versions of a sandwich in stock)
            18. + Produce exactly in the quantity demanded
         4. Examples of Demand Waiting for Supply
            1. Service Examples
            2. ER Wait Times: 58-year-old Michael Herrara of Dallas died of a heart attack after an estimated 19 hours in the local Hospital ER Some ER's now post expected wait times online / via Apps
            3. It takes typically 45 days do get approval on a mortgage; Strong link between wait times and conversion
            4. Waiting times for drive-through at McDonald's: 159 seconds; Long queues deter customers to join
            5. Production Examples
            6. Buying an Apple computer
            7. Buying a Dell computer
            8. => Make-to-order vs Make-to-Stock
      4. Five Reasons for Inventory
         1. Pipeline inventory: you will need some minimum inventory because of the flow time >0
         2. Seasonal inventory: driven by seasonal variation in demand and constant capacity
         3. Cycle inventory: economies of scale in production (purchasing drinks)
         4. Safety inventory: buffer against demand (Mc Donald's hamburgers)
         5. Decoupling inventory/ buffers: buffers between several internal steps
   5. Multiple flow units
      1. Approach 1: Adding-up Demand Streams
      2. Approach 2: A Generic Flow Unit ("Minute of Work")
      3. Steps for Basic Process Analysis with Multiple Types of Flow Units
         1. 1. For each resource, compute the number of minutes that the resource can produce
         2. 2. Create a process flow diagram, indicating how the flow units go through the process
         3. 3. Create a table indicating how much workload each flow unit is consuming at each resource
         4. 4. Add up the workload of each resource across all flow units.
         5. 5. Compute the implied utilization of each resource as
            1. The resource with the highest implied utilization is the bottleneck
            2. Note: you can also find the bottleneck based on calculating capacity for each step and then dividing the demand at this resource by the capacity
      4. Processes with Attrition Loss
         1. 1. Flow Units
         2. 2. Capacity
         3. 3. Implied Utilization
3. Productivity
   1. Introduction
      1. Basic definition of productivity
         1. Productivity = Units Output produced / Input used
         2. Example: Labor productivity Labor productivity = 4 units per labor hour (looks a lot like an processing time)
      2. Multifactor productivity
         1. Productivity = Output / (Capital$ + Labor$ + Materials$ + Services$ + Energy$)
      3. Waste and Inefficiencies
         1. Output: productive time; input: total time
         2. Some measures of productivity have natural limits (e.g. labor time, energy) What reduces productivity?
   2. Efficient Frontier
      1. There exists a tension between productivity and responsiveness
      2. Example: The US Airline Industry
   3. The Seven Sources of Waste
      1. Overproduction
         1. To produce sooner or in greater quantities than what customers demand
            1. Overproduced items need to be stored (inventory) and create further waste
            2. Bad for inventory turns
            3. Products become obsolete / get stolen / etc
         2. Examples
            1. 81.6 kg of food are trashed by the average German 61% of the trashing happens by households Large package sizes is the main reason
         3. Match Supply with Demand
      2. Transportation
         1. Unnecessary movement of parts or people between processes
            1. Example: Building a dining room and kitchen at opposite ends of a house, then keeping it that way
            2. Result of a poor system design and/or layout
            3. Can create handling damage and cause production delays
         2. Examples
            1. Crabs fished in the North Sea Shipped 2,500km South to Morocco Produced in Morocco Shipped back to Germany
         3. Relocate processes, then introduce standard sequences for transportation
      3. Rework
         1. Repetition or correction of a process
            1. Example: Returning a plate to the sink after it has been poorly washed
            2. Rework is failure to meet the "do it right the first time" expectation
            3. Can be caused by methods, materials, machines, or manpower
            4. Requires additional resources so that normal production is not disrupted
         2. Examples
            1. Readmissions to the ICU in a hospital (also called “Bounce backs”) Readmissions to the hospital after discharge (major component of Affordable Care Act)
         3. Analyze and solve root causes of rework => More in quality module
      4. Over-processing
         1. Processing beyond what the customer requires
            1. Example: Stirring a fully mixed cup of coffee
            2. May result from internal standards that do not reflect true customer requirements
            3. May be an undesirable effect of an operator's pride in his work
         2. Examples
            1. Keeping a patient in the hospital longer than what is medically required
         3. Provide clear, customer-driven standards for every process
      5. Motion
         1. Unnecessary movement of parts or people within a process
            1. Example: Locating (and keeping) a refrigerator outside the kitchen
            2. Result of a poor work station design/layout
            3. Focus on ergonomics
         2. Examples
            1. Ergonomics Look at great athletes
         3. Arrange people and parts around stations with work content that has been standardized to minimize motion
      6. Inventory
         1. Number of flow units in the system
            1. "Product has to flow like water"
            2. For physical products, categorized in: raw material, WIP, or finished products
            3. Increases inventory costs (bad for inventory turns)
            4. Increases wait time (see above) as well as the customer flow time
            5. Often times, requires substantial real estate
         2. => the BIGGEST form of waste
         3. Examples
            1. Loan applications at a bank
         4. Improve production control system and commit to reduce unnecessary "comfort stocks"
      7. Waiting
         1. Underutilizing people or parts while a process Examples completes a work cycle
            1. Example: Arriving an hour early for a meeting
            2. Labor utilization
            3. Idle time
            4. Note: Waiting can happen at the resource (idle time) But also at the customer level (long flow time)
         2. Examples
            1. Often, the time in the waiting room exceeds the treatment time by more than 5x
         3. Understand the drivers of waiting; more in Responsiveness module
      8. Wasteful vs Lean The IMVP Studies
         1. General Motors Framingham Assembly Plant Versus Toyota Takaoka Assembly Plant, 1986
         2. Gross assembly hours per car are calculated by dividing total hours of effort in the plant by the total number of cars produced
         3. Defects per car were estimated from the JD Power Initial Quality Survey for 1987
         4. Assembly Space per Car is square feet per vehicle per year, corrected for vehicle size
         5. Inventories of Parts are a rough average for major parts
      9. Understand Sources of Wasted Capacity
   4. Financial value of productivity
      1. Subway – Financial Importance of Operations
   5. KPI trees
      1. Subway – EBIT tree
         1. Profit
            1. Revenue
            2. Flow Rate
            3. Demand
            4. Min{}
            5. Capacity
            6. Station 1
            7. Station 2
            8. Station 3
            9. *
            10. $/customer
            11. -
            12. Cost
            13. Fix
            14. +
            15. Variable
            16. $/Sandwich
            17. Flow Rate
   6. OEE Framework / Quartile Analysis
      1. Overall Equipment Effectiveness
      2. OEE of an Aircraft
      3. Overall People Effectiveness
   7. Takt time
      1. Staffing / Capacity Sizing
         1. Typical situation in practice – Given are:
            1. Demand (forecasts)
            2. Activities that need to be completed
         2. Decision situation: how to build a staffing plan?
         3. Two strategies:
            1. Production smoothing (pre-produce)
            2. Staff to demand
      2. Line Balancing and Staffing to Demand
         1. => Staff to demand: start with the takt time and design the process from there
      3. What Do You Do When Demand Doubles?
         1. Ideal Case Scenario
      4. Balancing the Line
         1. Determine Takt time
            1. Assign tasks to resource so that total processing times < Takt time
            2. Make sure that all tasks are assigned
            3. => Minimize the number of people needed (maximize labor utilization)
         2. What happens to labor utilization as demand goes up?
            1. Difference between static and dynamic line balancing
      5. Line Balancing and Staffing to Demand
   8. Quartile analysis / Standardization
      1. Call Center Example
         1. Two calls to the call center of a big retail bank
         2. Both have the same objective (to make a deposit)
         3. Different operators
         4. Take out a stop watch
         5. Time what is going on in the calls.
      2. Beyond Labor Utilization: Quartile Analysis
         1. Biggest productivity differences for knowledge intense tasks
      3. Example: Emergency Department
         1. Analyzed data for over 100k patients in three hospitals
         2. 80 doctors and 109 nurses
         3. Up to 260% difference between the 10th %-tile and the 90th %-tile
         4. => Dramatic productivity effects
   9. Productivity Ratios
      1. Basic definitions of productivity
         1. Productivity = Output units produced / Input used
         2. Problems:
            1. Output is hard to measure=> often times, use revenue instead
            2. Multiple input factors (Labor, Material, Capital) => use one cost category
         3. Example:
            1. Labor productivity at US Airways 1995: Revenue: $6.98B Labor costs: $2.87B 2011: Revenue: $13.34B Labor costs: $2.41B
            2. Labor productivity at SouthWest 1995: Revenue: $2.87B Labor costs: $0.93B 2011: Revenue: $13.65B Labor costs: $4.18B
            3. Airline example: Revenue / labor costs = Revenue/RPM * RPM/ASM * ASM / Employee * Employees/Labor costs
            4. Labor Productivity Comparison between Southwest and US Airways
4. Customer Choice
   1. Introduction
      1. Forms of Variety
         1. Fit Based Variety
            1. Customers differ in shirt sizes
            2. Each customer has a unique utility maximizing shirt size
            3. The further you go away (in either direction) from that point, the lower the utility
            4. Hotelling's linear city
            5. Example: sizes, locations, arrival times
         2. Performance Based Variety
            1. Each customer prefers the high end model
            2. Customers differ in their valuation of quality (performance) and/or their ability to pay
            3. Vertical differentiation
            4. Example: screen resolutions, mpg, processor speeds, weight
         3. Taste Based Variety
            1. Customers differ in their preferences for taste
            2. Often times, these preferences vary over time
            3. Rugged landscape
            4. Example: taste for food, music, artists
      2. Economic Motives for Variety
         1. Heterogeneous preferences of customers
         2. Price discrimination
         3. Variety seeking by consumers
         4. Avoiding price competition in channel
         5. Channel self space
         6. Niche saturation and deterrence to market entry
   2. Impact on process capacity
      1. Ordering Custom Shirts
         1. Custom shirts ordered online
         2. Large variety of styles
         3. Basically infinitely many sizes
         4. Four weeks lead time
         5. Minimum order: 5 shirts
      2. Custom Tailored Shirts: Production Process
         1. Cutting Department
            1. The pattern is programmed into a machine and/or a cutting template is created. This takes a certain amount of set-up time IRRESPECTIVE of how many shirts will be produced afterwards.
         2. Sewing Department
            1. Sewing Section – Cut pieces of fabric are sewn together and inspected Assembly Section - Responsible for assembling shirts and measuring the size.
         3. Finishing Department
            1. Responsible for ironing shirts before folding, packaging and delivery to customers.
      3. Process Analysis with Batching
         1. Example: Cutting Machine for shirts 20 minute set-up time (irrespective of the number of shirts) 4 minute/unit cutting time 15 Shirts in a batch
         2. Capacity calculation for the resource with set-up changes:
            1. =15/(20min+15*4min/unit)=15/80 shirts/min
         3. Example Calculations
            1. What is the capacity of the cutting machine with a batch size of 15?
            2. Capacity of Cutting=15/(20+4*15)=15/80=0.188
            3. Capacity of Section1=8/40=0.2
            4. Capacity of Section2=5/30=0.167
            5. Capacity of Finishing=1/3=0.33
         4. Large Batches are a Form of Scale Economies
   3. Understanding the Diseconomies of Scale Extra inventory
      1. The Downside of Large Batches
         1. Large batch sizes lead to more inventory in the process
         2. This needs to be balanced with the need for capacity
         3. Implication: look at where in the process the set-up occurs
            1. If set-up occurs at non-bottleneck => decrease the batch size
            2. If set-up occurs at the bottleneck => increase the batch size
      2. General Definition of a Batch
         1. Product A: Demand is 100 units per hour Product B: Demand is 75 units per hour The production line can produce 300 units per hour of either product It takes 30 minutes to switch the production line from A to B (and from B to A) How would you set the batch size?
         2. p=1/300 h/unit Set time=0.5h
         3. AAAA S(AB) BBB S(BA) ...... 175=B/(Set time*2+B*p)=B/(1+B*1/300) B=420, B(A)=420*100/175=240, B(B)=180
      3. Introducing a Third Product into the Product Line
         1. Now, the Marketing folks of the company add a third product. Total demand stays the same. Product A1: Demand is 50 units per hour Product A2: Demand is 50 units per hour Product B: Demand is 75 units per hour How would you set the batch size?
         2. A1A1 S A2A2 S BBB S ...... 175=B/(Set time*3+B*p)=B/(1.5+B*1/300) B=630, B(A1)=630*50/175=180=B(A2), B(B)=270
   4. Choosing a good batch size
      1. Example Calculations
         1. Batch=5 Capacity of Cutting=5/20+4*5=0.125 (Bottleneck) Capacity of Section1=8/40=0.2 Capacity of Section2=5/30=0.167 Capacity of Finishing=1/3=0.333
         2. Batch=50 Capacity of Cutting=50/20+4*50=0.227 Capacity of Section1=8/40=0.2 Capacity of Section2=5/30=0.167 (Bottleneck) Capacity of Finishing=1/3=0.333
      2. How to Set the Batch Size – An Intuitive Example
      3. Process Analysis with Batching: Summary
         1. Batching is common in low volume / high variety operations
         2. Capacity calculation changes:
         3. This reflects economies of scale (similar to fix cost and variable cost)
         4. You improve the process by:
            1. Setting the batch size:
            2. (a) If set-up occurs at the bottleneck => Increase the batch size
            3. (b) If set-up occurs at a non-bottleneck => Reduce the batch size
            4. (c) Find the right batch size by solving equation
   5. Pooling Effects / Demand Fragmentation
      1. Variability of Demand / Polling
      2. Demand Fragmentation
         1. You have 3 products (different shirt sizes) Demand for each product could be 1, 2, or 3 with equal (1/3) probability How good is your forecast FOR YOUR OVERALL SALES?
   6. Building Flexibility: SMED / Heijunka
      1. The 6-stage SMED approach
         1. Reduce set-up so that you can change models as often as needed => Mixed model production (Heijunka)
      2. Flexibility vs Chaining
      3. Pooling vs Chaining
         1. Chaining is a form of partial flexibility ("pooling" light)
         2. Does not require full flexibility, but relies on a clever product-to-plant assignment
   7. Strategies to deal with variety / Investing in flexibility
   8. Limits to customization
5. Response Time
   1. Introduction
      1. Example
         1. Patients arrive, on average, every 5 minutes. It takes 10 minutes to serve a patient. Patients are willing to wait. What is the implied utilization of the barber shop? How long will patients have to wait?
            1. a=12 pat/h
            2. Cap=6 pat/h
            3. utilization=Demand/CAP=200%
         2. Patients arrive, on average, every 5 minutes. It takes 4 minutes to serve a patient. Patients are willing to wait. What is the implied utilization of the barber shop? How long will patients have to wait?
            1. a=12 pat/h
            2. CAP=15 pat/h
            3. utilization=Flow Rate/CAP=0.8
      2. A Somewhat Odd Service Process
      3. A More Realistic Service Process
      4. Variability Leads to Waiting Time
      5. The Curse of Variability - Summary
         1. Variability hurts flow
         2. With buffers: we see waiting times even though there exists excess capacity
         3. Variability is BAD and it does not average itself out
         4. New models are needed to understand these effects
   2. Waiting time models: The need for excess capacity
      1. Modeling Variability in Flow
      2. The Waiting Time Formula
         1. Waiting Time Formula
      3. Example: Walk-in Doc
         1. Newt Philly needs to get some medical advise. He knows that his Doc, Francoise, has a patient arrive every 30 minutes (with a standard deviation of 30 minutes). A typical consultation lasts 15 minutes (with a standard deviation of 15 minutes). The Doc has an open-access policy and does not offer appointments. If Newt walks into Francois’s practice at 10am, when can he expect to leave the practice again?
            1. p=15
            2. a=30
            3. u=p/a=0.5
            4. CVa=1
            5. CVp=1
            6. Tq=p*u/(1-u)*(CVa^2+CVp^2)/2=15
            7. T=Tq+p=30
      4. Summary
         1. Even though the utilization of a process might be less than 100%, it might still require long customer wait time
         2. Variability is the root cause for this effect
         3. As utilization approaches 100%, you will see a very steep increase in the wait time
         4. If you want fast service, you will have to hold excess capacity
   3. More on Waiting time models / Staffing to Demand
      1. Waiting Time Formula for Multiple, Parallel Resources
         1. Waiting Time Formula for Multiple (m) Servers
      2. Example: Online retailer
         1. Customers send emails to a help desk of an online retailer every 2 minutes, on average, and the standard deviation of the inter-arrival time is also 2 minutes. The online retailer has three employees answering emails. It takes on average 4 minutes to write a response email. The standard deviation of the service times is 2 minutes. Estimate the average customer wait before being served.
            1. p=4, a=2, m=3
            2. u=p/am=0.6667
            3. CVa=1, CVp=0.5
            4. Tq=1.19min
      3. Summary of Queuing Analysis
         1. Utilization (Note: make sure <1)
         2. Time related measures
         3. Inventory related measures (Flow rate=1/a)
      4. Staffing Decision
         1. Customers send emails to a help desk of an online retailer every 2 minutes, on average, and the standard deviation of the inter-arrival time is also 2 minutes. The online retailer has three employees answering emails. It takes on average 4 minutes to write a response email. The standard deviation of the service times is 2 minutes. How many employees would we have to add to get the average wait time reduced to x minutes?
      5. What to Do With Seasonal Data
      6. Service Levels in Waiting Systems
         1. Target Wait Time (TWT)
         2. Service Level = Probability{Waiting Time≤TWT}
         3. Example: Big Call Center
            1. starting point / diagnostic: 30% of calls answered within 20 seconds
            2. target: 80% of calls answered within 20 seconds
   4. Capacity Pooling
      1. Managerial Responses to Variability: Pooling
         1. Independent Resources 2x(m=1)
            1. Example
            2. Processing time=4 minutes
            3. Inter-arrival time=5 minutes (at each server)
            4. m=1, Cva=CVp=1
            5. u = 0.8, Tq = 16
         2. Pooled Resources (m=2)
            1. Example
            2. Processing time=4 minutes
            3. Inter-arrival time=2.5 minutes
            4. m=2, Cva=CVp=1
            5. u = 0.8, Tq = 7.24
      2. Pooling: Shifting the Efficient Frontier
      3. Limitations of Pooling
         1. Assumes flexibility
         2. Increases complexity of work-flow
         3. Can increase the variability of service time
         4. Interrupts the relationship with the customer / one-face-to-the-customer
   5. Scheduling / Access
      1. Managerial Responses to Variability: Priority Rules in Waiting Time Systems
         1. Flow units are sequenced in the waiting area (triage step)
         2. Provides an opportunity for us to move some units forwards and some backwards
         3. First-Come-First-Serve
            1. easy to implement
            2. perceived fairness
            3. lowest variance of waiting time
         4. Sequence based on importance
            1. emergency cases
            2. identifying profitable flow units
         5. Shortest Processing Time Rule
            1. Minimizes average waiting time
            2. Problem of having “true” processing times
      2. Appointments
         1. Open Access
         2. Appointment systems
   6. Redesign the Service Process
      1. Reasons for Long Response Times (And Potential Improvement Strategies)
         1. Insufficient capacity on a permanent basis => Understand what keeps the capacity low
         2. Demand fluctuation and temporal capacity shortfalls Unpredictable wait times => Extra capacity / Reduce variability in demand Predictable wait times => Staff to demand / Takt time
         3. Long wait times because of low priority => Align priorities with customer value
         4. Many steps in the process / poor internal process flow (often driven by handoffs and rework loops) => Redesign the service process
      2. The Customer's Perspective
         1. Two types of wasted time:
            1. Auxiliary activities required to get to value add activities (result of process location / lay-out)
            2. Wait time (result of bottlenecks / insufficient capacity)
      3. Process Mapping / Service Blue Prints
         1. How to Redesign a Service Process
            1. Move work off the stage Example: online check-in at an airport
            2. Reduce customer actions / rely on support processes Example: checking in at a doctor's office
            3. Instead of optimizing the capacity of a resource, try to eliminate the step altogether Example: Hertz Gold – Check-in offers no value; go directly to the car
            4. Avoid fragmentation of work due to specialization / narrow job responsibilities Example: Loan processing / hospital ward
            5. If customers are likely to leave the process because of long wait times, have the wait occur later in the process / re-sequence the activities Example: Starbucks – Pay early, then wait for the coffee
            6. Have the waiting occur outside of a line Example: Restaurants in a shopping malls using buzzers Example: Appointment
            7. Communicate the wait time with the customer (set expectations) Example: Disney
   7. Loss Models
      1. Different Models of Variability
         1. Waiting problems
         2. Utilization has to be less than 100% Impact of variability is on Flow Time
         3. Loss problems
         4. Demand can be bigger than capacity Impact of variability is on Flow Rate
         5. Variability is always bad – you pay through lower flow rate and/or longer flow time
         6. Buffer or suffer: if you are willing to tolerate waiting, you don't have to give up on flow rate
      2. Analyzing Loss Systems
         1. Finding Pm(r)
            1. Define r = p / a
            2. Example: r=2 hours/3 hours=0.67
            3. Recall m=3
            4. Use Erlang Loss Table
            5. Find that P3 (0.67)=0.0255
            6. Given Pm(r) we can compute:
            7. Time per day that system has to deny access
            8. Flow units lost = 1/a * Pm (r)
      3. Implied utilization vs probability of having all servers utilized: Pooling Revisited
      4. Erlang Loss Table
6. Quality
   1. Introduction
      1. Two dimensions of quality
         1. conformance
         2. performance
      2. Assembly Line Defects
         1. = (1-0.01)^9=0.9135
      3. The Duke Transplant Tragedy
         1. 17 year old Jesica Santillan died following an organ transplant (heart+lung) Mismatch in blood type between the donor and Jesica Experienced surgeon, high reputation health system About one dozen care givers did not notice the mismatch The offering organization did not check, as they had contacted the surgeon with another recipient in mind The surgeon did not check and assumed the organization offering the organ had checked It was the middle of the night / enormous time pressure / aggressive time line
         2. A system of redundant checks was in place A single mistake would have been caught But if a number of problems coincided, the outcome could be tragic
      4. Swiss Cheese Model
         1. =0.01^3
      5. The Nature of Defects
         1. Assembly line example: ONE thing goes wrong and the unit is defective
         2. Swiss cheese situations: ALL things have to go wrong to lead to a fatal outcome
         3. Compute overall defect probability / process yieldWhen improving the process, don't just go after the bad outcomes, but also after the internal process variation (near misses)
   2. Defects / impact on flow
      1. CAP1=1/5 units/min=12 units/h, CAP2=1/4 units/min=15 units/h, CAP3=1/6 units/min=10 min/h
      2. Demand1=2D, Demand2=2D, Demand3=D
      3. Impl Util1=2D/12, Impl Util2=2D/15, Impl Util=10/D
      4. Processing Times2=0.7*4+0.3*(4+4)=5.2
      5. CAP1=1/5 units/min, CAP2=1/5.2 units/min, CAP3=1/2 units/min
      6. Demand1=D, Demand2=1.3D, Demand3=D
      7. Impl Util1=5D, Impl Util2=6.76D, Impl Util=2D
      8. Impact of Defects on Variability: Buffer or Suffer
         1. Processing time of 5 min/unit at each resource (perfect balance) With a probability of 50%, there is a defect at either resource and it takes 5 extra min/unit at the resource to rework => What is the expected flow rate?
         2. 1/10
      9. The Impact of Inventory on Quality
         1. Inventory takes pressure off the resources (they feel buffered): demonstrated behavioral effects Expose problems instead of hiding them
      10. Operations of a Kanban System: Demand Pull
          1. Visual way to implement a pull system Amount of WIP is determined by number of cards Kanban = Sign board Work needs to be authorized by demand
      11. Ishikawa Diagram
   3. Six sigma and process capability
      1. M&M Exercise
         1. A bag of M&M's should be between 48 and 52g
         2. Measure the samples on your table: Measure x1, x2, x3, x4, x5 Compute the mean (x-bar) and the standard deviation Number of defects
         3. All data will be compiled in master spread sheet Yield = %tage of units according to specifications How many defects will we have in 1MM bags?
      2. Measure Process Capability: Quantifying the Common Cause Variation
      3. The Concept of Consistency: Who is the Better Target Shooter?
   4. Two types of variation
      1. Detect Abnormal Variation in the Process: Identifying Assignable Causes
         1. Track process parameter over time
            1. average weight of 5 bags
            2. control limits
            3. different from specification limits
         2. Distinguish between
            1. common cause variation (within control limits)
            2. assignable cause variation (outside control limits)
      2. Statistical Process Control
         1. Capability analysis
            1. What is the currently "inherent" capability of my process when it is "in control"?
         2. Conformance analysis
            1. SPC charts identify when control has likely been lost and assignable cause variation has occurred
         3. Investigate for assignable cause
            1. Find "Root Cause(s)" of Potential Loss of Statistical Control
         4. Eliminate or replicate assignable cause
            1. Need Corrective Action To Move Forward
   5. Detect / Stop / Alert
      1. Information Turnaround Time
         1. Assume a 1 minute processing time
         2. Inventory leads to a longer ITAT (Information turnaround time) => slow feed-back and no learning
      2. Cost of a Defect: Catching Defects Before the Bottleneck
      3. Detecting Abnormal Variation in the Process at Toyota: Detect – Stop - Alert
         1. Jidoka
            1. If equipment malfunctions / gets out of control, it shuts itself down automatically to prevent further damage
            2. Requires the following steps:
            3. Detect
            4. Alert
            5. Stop
         2. Andon Board / Cord
            1. A way to implement Jidoka in an assembly line
            2. Make defects visibly stand out
            3. Once worker observes a defect, he shuts down the line by pulling the andon / cord
            4. The station number appears on the andon board
      4. Two (similar) Frameworks for Managing Quality
         1. Some commonalities:
            1. Avoid defects by keeping variation out of the process
            2. If there is variation, create an alarm and trigger process improvement actions
            3. The process is never perfect – you keep on repeating these cycles
   6. Problem solve / improve
      1. Root Cause Problem Solving
         1. Ishikawa Diagram
            1. A brainstorming technique of what might have contributed to a problem
            2. Shaped like a fish-bone
            3. Easy to use
         2. Pareto Chart
            1. Maps out the assignable causes of a problem in the categories of the Ishikawa diagram
            2. Order root causes in decreasing order of frequency of occurrence
            3. 80-20 logic
   7. Lean Operations
      1. The Ford Production System
         1. Influenced by Taylor; optimization of work
         2. The moving line / big machinery => focus on utilization
         3. Huge batches / long production runs; low variety
         4. Produced millions of cars even before WW2
         5. Model built around economies of scale => Vehicles became affordable to the middle class
      2. The Toyota Production System
         1. Toyota started as a maker of automated looms
         2. Started vehicle production just before WW2
         3. No domestic market, especially following WW2
         4. Tried to replicate the Ford model (produced about 10k vehicles)
         5. No success due to the lack of scale
         6. Around 1950, TPS was born and refined over the next 30 years => Systematic elimination of waste => Operating system built around serving demand
      3. Toyota Production System: An Overview
      4. The Three Enemies of Operations