Process Analysis and Optimization in Business Operations
Class 1: Introduction to Process Analysis
Key Concepts:
- Business Process: Transformation of inputs into outputs through a network of activities and buffers, utilizing resources, IT, and management.
- 4 Key Elements of a Business Process: Inputs, outputs, transformation, resources (capital, labor).
- Customer Value Dimensions: Time, price, quality, variability.
- Process Competencies: Flow time/cycle time, cost, quality, flexibility.
- Aligned Strategy: Variability output is aligned with flexibility process.
- Competitive Product Space Diagram: Operations frontier (line curve), world class (closest to line curve), dominated (near line curve), improvement (shift up/closer to line curve), tradeoff, tradeoff (shift to left/right along line curve).
- Life Trajectory of a Successful Commercial Product: Straight line with decreasing slope starts from job shop and ends at assembly line/flow shop.
Class 2: Process Architecture
Key Concepts:
- Operational Managers: Take a process view of business and other organizations. Focus on transformation.
- Operational Effectiveness: Firms need to develop operational capabilities, i.e., process competencies.
- Proper Operations Strategies: Provide compatibility between firm’s business strategy and its business process.
Class 3: Process Analysis
Process Architecture 1: Types of Processes
- Job Shop: Low product volume, low equipment dedication, high product variety, high machine setup frequency, high labor skills, high variable cost.
- Manufacturing Cell: Medium product volume, medium equipment dedication, medium product variety, medium machine setup frequency, medium labor skills, medium variable cost.
- Assembly Line: High product volume, high equipment dedication, low product variety, low machine setup frequency, low labor skills, low variable cost.
Process Architecture 2: Product-Process Matrix
- Job Shop: Low volume, low standardization, one of a kind products. EX: Boeing Aircraft, Steinway Piano, J. Lo’s Wedding Dresses, commercial printer, architecture firm.
- Batch: Low volume, many products. EX: Caterpillar Truck, Tank, Scholars Inn Bakehouse Bread, heavy equipment, auto repair.
- Assembly Line/Flow Shop: Higher volume, few major products. EX: Laptop, Toyota Camry, Coca Cola Bottling, auto assembly, auto lube shop.
- Continuous Flow: High volume, high standardization, commodity products. EX: oil refinery.
Key Steps in Process Analysis:
- Process Flow Diagram: Define the process.
- Bottleneck Analysis: Determine the capacity of each resource and of the process.
Bottleneck Analysis:
- Bottleneck: The slowest part (resource) of the process. It determines the cycle time of the process.
- Cycle Time of the Process = Cycle Time of the Bottleneck
- Capacity/Throughput Rate of the Process = Capacity/Throughput Rate of the Bottleneck
Cycle Time (CT):
- The average time between completion of successive units.
- Bottleneck would determine the whole process cycle time.
- EX: CT = 15s/unit.
Capacity (Rate):
- The MAXIMUM rate that the units flow through the system.
- Capacity = Total available time/Cycle time
- Bottleneck would determine the whole process capacity.
- EX: capacity rate = 4 units/min.
Flow Time:
- Total length of time a unit spends in the system
- EX: FT = 25s/unit.
Buffering:
- Storage area between stages where output of a stage is placed there before being used in next stage.
- Allows stages to operate independently.
- Prevents blocking and starving.
Blocking:
- Occurs when activities in a stage must stop because there is no place to deposit just completed items.
- EX: If there is gap in production in which Station 2 (fast) having to wait for Station 1 (slow), this is considered blocking.
Starving:
- Occurs when activities in a stage must stop because there is no work.
- EX: If there is gap in production in which Station 1 (fast) having to wait for Station 2 (slow), this is considered starving.
Class 4: Kristen’s Cookie Company
Key Lessons Learned:
- Improving the non-bottleneck activity does not change the capacity of the whole process. However, it may change the flow time.
- Improving on bottleneck can improve the capacity rate. The new capacity rate of the system depends on the capacity of the new bottleneck.
- Improving the capacity rate of the bottleneck may or may not improve flow time, depending on how the bottleneck is improved.
Class 5: Capacity Analysis
Six Levers for Increasing Capacity:
- Investment in more labor and/or machines.
- Reduce the activity time on all bottleneck resources.
- Increase the availability of all bottleneck resources.
- Make activities self-service.
- Shift work from a bottleneck to a non-bottleneck.
- Combine and parallelize.
Key Concepts:
- Demand Rate = Inflow Rate; Throughput Rate = Output Rate.
- Throughput Rate: How many flow units pass through the process per unit of time (output rate of the process).
- (Theoretical) Utilization = Demand Rate/Capacity Rate.
- Actual Utilization = Min (100%, Demand Rate/Capacity Rate).
- Critical Path Analysis: Critical path is the sequence of steps through which each customer must be processed, that takes the longest to complete. => Flow time of the process = flow time of the critical path.
- Critical Path vs Bottleneck:
- Critical Path: Determines flow time; To shorten flow time, focus on critical path; Watch out for new critical paths to emerge as you speed up the critical activities.
- Critical Path: Determines process capacity; To increase process capacity , focus on bottleneck; Watch out for new bottlenecks to emerge as you speed up the bottleneck.
EX of a Parallel Process:
- Step 1 requires Brew with CT = 15s and Froth with CT = 20s while Step 2 requires Pour with CT = 5s.
- Process cycle time = 20s/cup (froth is bottleneck).
- Flow time = max(15, 20) + 5 = 25s/cup.
- Capacity = 60s/cycle time = 3 cups/min.
Class 6: Inventory Buildup Diagram
Little’s Law: Inventory = Throughput Rate * Flow Time
(I = R * T).
+ Throughput: Rate at which jobs “come and go” through the process (jobs/time = 1/Cycle Time). EX: A box arrives at the system every 30 seconds. => R = 2 boxes/min.
+ Flow time: Time a job spends in a process (time). EX: A box takes 3 minutes to go through the system. => T = 30s.
+ Inventory: Jobs that accumulate in a process (jobs). EX: On average how many boxes in the system? => I = R * T = 6 boxes.
+ Capacity: Maximum achievable average throughput (jobs/time).
– Inventory Buildup: Time-varying Demand
1. Demand Rate No inventory build-up.
2. Demand Rate Reduces backlog rate = Capacity Rate – Demand Rate.
3. Demand Rate > Capacity Rate => Inventory buildup rate = Demand Rate – Capacity Rate.
– Ri : inflow rate; Ro: outflow rate; △I: inventory buildup rate; (△I = Ri – Ro).
Class 7) Production Planning
– The ABC’s of Optimization Modeling:
+ “Adjust” (decision variables): The choices the manager is able to change
+ “Best”: (objective function): The manager’s goal, expressed mathematically
+ “Constraints”: Limitations that restrict the manager’s choices
– Three Steps to Building an Optimization Model: 1) Build a conceptual model (Create model in words); 2) Build a mathematical model (create model using math equations); 3) Build a spreadsheet model (create model in Excel).
– EX of Conceptual Model:
+ “Adjust” (decision variables): # of tablets H and Z to product every day
+ “Best”: (objective function): To maximize the daily profit
+ “Constraints”:
• Demand constraint: # of tablets produced does not exceed the daily demand
• Capacity constraint (for each resource): # of hours used does not exceed the available hours
• Non-negative constraint: # of tablets produced >= 0
– EX of Mathematical Model:
+ “Adjust” (decision variables):
• nH : number of tablets H to produce
• nZ : number of tablets Z to produce
+ “Best”: (objective function): 20nH + 30nZ
+ “Constraints”:
• Demand constraint: nH ≤ 100, nZ ≤ 40
• Capacity constraint: nH ≤ 70 , nZ ≤ 50 , nH + 2nZ ≤ 120 , nH + nZ ≤ 96
• Non-negative constraint: nH ≥ 0, nZ ≥ 0
– Answer Report: contains a basic summary of the objective function, decision variables, and constraints, including their cell references and final values.
– Sensitivity Report: allows you to analyze how the optimal solution is affected by changes, within specified ranges, in the objective coefficients and the right-hand side values of the constraints (more details below).
Senstitivity Report)
– Show info about the objective function: • its optimal value • coefficient ranges (ranges of optimality)
– Show info about the decision variables: • their optimal values • their reduced costs
– Show info Information about the constraints: • the amount of slack or surplus • the shadow prices • right-hand side ranges (ranges of feasibility)
– Top Half of Report (Variable Cells): Changes in price/contribution margin
– Bottom Half of Report (Constraints): Changes in capacity of the resources
– Reduced Cost (Variable Cells):
+ For any decision variable at its upper (lower) bound in the optimal solution, the Reduced Cost indicates how much the objective function improves (gets worse) when this bound is relaxed (tightened) by one unit.
+ Interpretation: The upper bound of the decision value in the production planning problem is the demand of each product.
+ Reduced cost = how much the profit increases if the demand increases by 1 unit.
– Range of Optimality (Variable Cells):
+ The Allowable Increase and Allowable Decrease for the decision variables indicate how much you can change the objective coefficient (i.e. contribution margin) before the optimal solution would be different.
+ The optimal solution does not change if the change of the contribution margin is within the range (if there is only one change at a time). But the profit does change!
– Shadow Price (Constraints):
+ The Shadow Price indicates the change in the optimal value of objective when the right-side of a constraint changes by one unit.
+ The objective function improves (gets worse) by the value of shadow price when a constraint is relaxed (tightened) for one unit.
+ Resource A (de)increases by 1 => Profit (de)increases by shadow price of A.
– Range of Feasibility (Constraints):
+ The Allowable Increase and Allowable Decrease for the constraints indicate how much you can change the right-hand side of the constraint before the shadow price given by the sensitivity report is no longer valid.
+ Shadow price remains the same if the change is within the range (if there is only one change at a time).
– Simultaneous Changes: 100% rule
+ Simultaneous changes in objective coefficients (constraint right-hand sides) will not change the optimal solution (shadow prices) as long as the sum of the percentages of the change divided by the corresponding maximum allowable change in the range of optimality (range of feasibility) for each coefficient (right-hand side) does not exceed 100%.
+ 100% rule does not say, however, that the optimal solution (shadow prices) changes when the sum exceeds 100%.
+ It applies to both the coefficient (the contribution margin in our setting) and the constraint (the capacity in our setting).
1. Check for each change: Relative change = Proposed change/Allowable increase(decrease).
2. Sum up all the relative changes
a) If the sum > 100% => We cannot use the sensitivity report to compute the profit. Need to rerun-Solver.
b) If the sum Compute the new profit using the sensitivity report.
– How to Read the Sensitivity Report:
+ Changes in Objective Coefficient (Contribution Margin). => Is the change within the allowable range?
• Yes. => The solution remains optimal. Update the profit using the new contribution margin.
• No. => The optimal solution changes. Need to re-run Solver to obtain the new optimal product mix.
+ Changes in Available Resources. => Is the change within the allowable range?
• Yes. => Shadow price = marginal value of the one unit of the resource. Shadow prices remain unchanged within the allowable range.
• No. => The shadow price changes if the change in resource capacities exceeds the range.
+ If multiple changes happen simultaneously, use 100% rule to check the allowable range.
