Process Improvement and Inventory Management in Operations

Project Management

Process Capability

Process capability is a measure of how many products produced by a process are defective.

  1. A process with a process capability of 4-sigma is better than 2 sigma.
  2. The higher the variability of a process, the lower its process capability.
  3. The choice of a product’s design specifications affects process capability even if the mean and standard deviation of the process do not change.
  4. Process capability generally decreases as the design specifications tighten (i.e., the distance between the upper and lower design specifications decreases), assuming the process variability remains the same.
  5. Customer wants are not always adequately reflected in product specifications.

Process Improvement

Process improvement depends upon a series of small experiments to increase learning about the process, not all of which have to be successful.

  1. Continuous process improvement is a series of incremental adjustments to the process rather than redesigning the process from the “ground-up.
  2. Continuous process improvement ideally involves a systematic and ongoing approach rather than sporadic, ad hoc efforts.


Quality depends on the expectations and wants of the customers. Quality includes:

  • Reliability: Works without problems as perceived by the customer.
  • Functionality: When the product is working, how well does it satisfy the customer? Satisfaction usually depends on multiple product features such as performance, aesthetics, even snob appeal.
  • Conformance: Essentially process capability. Warning: specification limits may not match customer wants!

SCOM factors can impact all of these.

Quality is measured by comparing customer wants and process output.

Customer wants → design specifications (upper and lower specification limits)
– Tied tightly to new product development
Process output → process mean and variability
– Standard Deviation, also known as “sigma” or s, is the most common measure of variability in operations.

Process Capability increases with the amount of process output that falls within the design specifications. Process output that falls outside are Defects.

To get the Process Capability, we measure, in terms of std. deviations (sigmas), how close the mean is to the nearest (customer-determined) specification limit, or more formally:


This is a picture of a 3-sigma quality process, you want to target a 6 sigma quality. 6-s quality: (~3.4 defects per million)
Defect rates would be cut by ~2900 times over 99% and ~800 times over 3s.

Expert-driven, Benchmarking, Prioritization of PI efforts

Four Fundamental Principles of Quality Improvement

  1. Focus on customer satisfaction
  2. Focus on Process
  3. Continuous improvement
  4. Participation by everyone

The DMAIC Approach to Process Improvement (PI)


P4SkfcxTrekAAAAASUVORK5CYII= Black Belt PI tools → 1) process flow diagram (define) BCMRv8Bp9JV2wkwEZEAAAAASUVORK5CYII= mANE6DjHxUXoTI6L0JkcF6EzOS5CZ3JchM7kuAidyXEROpPjInQmx0XoTE4nQseZDBehMzlBhI7jOI7jOI7jOI7jOI7jOI4zGbPZf83rSIatweM3AAAAAElFTkSuQmCC

2) Fishbone Ishikawa Diagram (analyze) 3) Pareto Chart (analyze) 4) Histogram 5) Run chart 6) Correlation

Improve: 7) Implement Lean- Reduce waste, flowtimes; Visual management, etc. 8) Benchmarking- Trick is to benchmark outside the box! 9) Control Process Control Charts

Key Takeaways

  • Variability is evil and raises defect levels.
  • Quality is defined by the customer and may vary by customer.
  • Process improvement efforts can not only improve the product’s quality but can also increase capacity and reduce waste and costs.
  • Processes can always be improved, hence the need for continuous improvement. These improvements may often be small. Not all of these efforts may be successful.

Managing Inventory

What is a Project?

A project is “a number of activities that result in a one-of-a-kind product or service.”

Measuring Project Success

How do we measure a project’s success? Timing, Cost (Budget), Quality (Scope): Not just absolute amounts, but also relative to planning estimates?

But 75% of all projects are late, over-budget, or both!

Top 10 Reasons for Schedule Slips and Budget Overruns

  1. Insufficient front-end planning
  2. Unrealistic project plan
  3. Project scope underestimated
  4. Customer/management changes
  5. Insufficient contingency planning
  6. Inability to track progress
  7. Inability to detect problems early
  8. Insufficient number of checkpoints
  9. Staffing problems
  10. Technical complexities

Project Management Tools

  • Work Breakdown Structure (WBS): Explains exactly what needs to be done.
  • The Gantt Chart: Simple and easy to explain to the uninitiated.
  • Critical Path Method (CPM): Explicitly represents dependencies. Can calculate time-cost tradeoffs.

Performing a Forward Path Analysis

Earliest start (ES) is 0 for activities with no predecessor.
– calculate ES and EF (earliest finish) for each activity.

Performing a Backward Path Analysis

LF (latest finish) is EF for activities with no successor. Calculate LS and LF (latest finish) for each activity.
Establish slacks (time you can delay activity without violating the latest finish i.e., LF-EF) and the critical path (path that determines the time of the project) slack= LS-ES.

8DcupTlOSSCSEAAAAASUVORK5CYII= x9DBmzNWFUogQAAAABJRU5ErkJggg== critical path = longest duration


Factors to Consider When Crashing a Project

  • The amount by which an activity is crashed is, in fact, permissible.
  • Taken together, the shortened activity durations will enable us to finish the project by the due date or sooner.
  • The total cost of crashing is as small as possible.

If there is only one critical path, then select the activity on this critical path that (a) can still be crashed, and (b) has the smallest crash cost per period.
If there is more than one critical path, then select one activity from each critical path such that (a) each selected activity can still be crashed, (b) the total crash cost of all selected activities is the smallest. Note that the same activity may be common to more than one critical path.


Purchasing is the single biggest cost for most businesses, accounting for about 60% of the average company’s total cost.

Sourcing is crucial for Services (data processing, customer support, health, law, .) as well.

Spectrum of Sourcing Relationships

Buy = Outsource Make = Insource
Market Buy Long Term Relationship
increased integration → – Ford used to own rubber plantations in Brazil, iron mines in the US, ships, … (vertical integration)

  • Core competencies – What you do well – In-Firm Knowledge
  • Strategic Sourcing – Get experience and knowledge that you don’t have. – Typically lower initial cost.

Unintended Consequence- Lose Product Knowledge – When IBM outsourced the PC processor and the OS system in the 1980s, did it foresee the rise of Intel and Microsoft?


New Balance Athletic Shoe, Inc. – August 2005

Founded in 1906 as New Balance Arch, Arch supports and prescription footwear were cornerstones till 1961. Launched the world’s first performance running shoe made with a ripple sole and in multiple widths. Was primarily a mail-order business in 1972 when Jim Davis bought the company and committed himself to uphold the company’s values of fit, performance, and manufacturing (customer’s consumption utility, Session 1, Slide 6).
First international sales office and manufacturing operations in 1978. Debuted the 990 running shoe in 1982 – the first athletic shoe to reach a price of $100.

New Balance’s Philosophy

  • We do not endorse athletes.
  • Every one of our shoes is a performance product.
  • We sell every shoe that we make in multiple widths. – a 13EEEE size customer will go back to the same retailer.
  • Being private we can act more nimbly and be more socially responsible.
  • We carry enough inventories so that our retailers can get continuous fill-ins and operate with low inventories. – NB enables retailers to build customer loyalty. – NB has far greater consumer loyalty than its competitors. – Inventories ↓ Markdown Risks ↓ Price Support ↑
  • We have domestic factories. – We can maintain manufacturing capabilities. – Our delivery lead times are shorter. We can respond faster. It supports continuous fill-ins by retailers. – “What we learn by doing things domestically, we share with our partners abroad.”
  • We are manufacturers. And we are mediocre marketers by design. – “We tried a couple of times …. and we failed, drastically. So then I woke up one morning and I said, ‘We’re not any good at that. We are really good at this.’ So we concentrated on doing this instead of that.” – Jim Davis LASTLY NB’s marketing spend is much lower than that of its competitors.
  • We are very team-oriented and we empower people.
  • Longstanding commitment to social responsibility.- “made people feel good about dealing with the company.”
  • The company’s culture is very entrepreneurial.
  • NB’s US workforce is nonunionized. – “wouldn’t be able to do that if we were unionized.”
  • Factory employees are one of the greatest forces for change in the company.
  • “Endorsed By No One”; “For Love or Money.” – “everyday athletes playing for the love of the game.”

New Balance’s Sourcing Strategy

  • What has New Balance outsourced and offshored?
    – High volume products with predictable order patterns and lower margins – lower needs for width sizing
  • What has New Balance kept “in house” and where? – higher margin models requiring a – wider variety of sizes and widths – on short notice – materials warehouses and manufacturing plants in Maine and Massachusetts

mQycfvrpwZyjTzn1Zfny5dXOnTs7f81WCCGEEELMDXqTIOYc3hDwtmBiYqJ66623wteK7YvFmHHVBEEIIYQQYlSoqv8fF4BrpNxehPwAAAAASUVORK5CYII= New Balance Update – 2010:


Inventory Management

Let D be the avg. demand per unit time also ROP = average demand * average delivery Leadtime
let H be Holding cost per unit/year & let S be the ordering cost (incl. shipping)

Q* = Optimal Order Quantity (EOQ) Total Cost = Order Cost (Setup Cost) + Holding Cost
Set up Cost = Number of Orders placed per year * S Number of Order per Year =Annual Demand D/*Q
S = Set up Cost per Order Holding Costs = Average Inventory * hc

Average Inventory =. Q* /2 Annual Ordering Cost = orders per year * ordering cost per year

Example: d= 2,000 per Week. D = 2,000* 52 = 104,000 per year → Q* = 5,551 packages
S = $80+1*$20 = $100 → Average Inventory. = Q*/2. = 2,775.5 & Number of Orders per Year. =. D/Q = 18.74

H = .3* 2.25 per package → the .3 is 30% and the 2.25 is cost per package

The general form for SS is given by assuming a normal demand distribution.→ SS= Z*Standard deviation of demand * Sqr root of lead-time

where z depends on the target service level. The service level is the probability of not stocking out.

Inventory holding costs increase by warehouse costs increasing

The Bullwhip Effect

There is a tendency for variability to escalate with each stage in the distribution process: usually from a corresponding degradation in the quality of demand information.

Delivery delays, demand forecast updating, order batching, price fluctuations (Price Promotions), shortage gaming

Optimal ordering quantity increases: holding rates decrease, ordering costs increase, shipping costs increase, the purchase cost of an inventory item decreases,

MT tool and Vehicles are correlated → WFETier 2 supplier revenue spike 1997, Retails sales to customers and orders received by manufacturers spiked like crazy in 2019, same with semiconductors being aligned with industrial production

Vertically integrating, increasing visibility into the inventory, reducing delivery lead times

Information sharing (POS info helps here)- Everyone in the chain knows consumer demand and each other’s inventories= (Delivery Delays, Demand forecast updating, Shortage gaming.)- channel alignment (reduce suppliers)= (Delivery delays, shortage gaming)- every-day low prices, long-term contracts (Price fluctuations)- Reduce Batching – operational efficiency – Increase deliveries. Reduce delivery delays and inventories. (Delivery delays, demand forecast updating, batching)

Semester Takeaways

  • Choosing the right process design is critical to success.
  • You will always have a bottleneck.
  • You are only as good as the weakest link of your supply chain (Process).
  • Variability escalates as you move down the supply chain (bullwhip).
  • When establishing an inventory policy you want to minimize cost.
  • Your process design determines the amount of time spent waiting.
  • Proper management is critical to the timing and cost of projects.
  • The definition of quality comes from the customer, customer dissatisfaction= gap from perceptions to expectations.
  • Most firms spend too little on up-front planning.

EOQ- Continuous Process Improvement

  • Defining the goal of a process improvement initiative.
  • Gathering process data, creating a process flow diagram, focusing on determining quality by the customer.

Advantages of Inventory

  • Increase sales.
  • Decrease order-to-delivery cycle time (lead time).
  • Decrease Stockouts.
  • Improve Utilization (Get the most out of your Capacity).
  • Protects against blockage and starvation.
  • Decouples downtimes of different operations.

Each operation operates at a 95% uptime (5% downtime). What is the system uptime?

Uptime without inventory = .95^3 = 85.7%

Old Unit Stuff

For capacity → 2 * 1/ (14 mins/cust) * 60 mins/hr = 8.57 custs/hr

L=λ×WL is the average number of customers in the system (waiting room)
λ is the average throughput rate W is the average time a customer spends in the system

ROFT =5m+14m+6m+max(4m+8m, 5m)+4m = 41 minutes

Little’s Law= Average Inventory = Average Flow time (FT) x Average Flow Rate (FR)

Throughput = MIN(arrival rate, Process Cycle) is worded like # of patients seen per hour is on average 5 cust/hours – FLOWRATE AND THRUPUT ARE SYNONYMS

Inventory = if there are average 4 patients in the office at one time

Average Flow Rate = I/FT & Average Flow Time= 1/FR

Bottleneck= smallest capacity

Customers arrive every 6 mins = 1/6 x 60

Utilization = what you use/ what you pay for = throughput/capacity =

Cycle time/ TAKT time = 4 customers arrive on average / bottleneck

Cycle time = 1/capacity TAKT time > cycle time

Takt time – 1/throughput

Process cycle efficiency = rush order flow time/flowtime (average member visiting)

Profit per hour = revenue per hour – material costs per hour – labor costs/hour – fixed cost/ hour

Profit as a % of revenue = Profit per hour / revenue per hour (process capacity x each client pays $$

Operating Leverage = (P-VC)/ ((P-VC) – (F/Q))

Breakeven Volume = FC/ (P-VC)

% increase in profit = operating leverage x % increase in demand x 100

Operation Leverage = (P-VC) x Q / (P-VC) x Q – Total Fixed Cost

The customer is not a resource and Eco was not the bottleneck, so the process capacity stays the same

Breakeven = FC / (P-V)

Increase in Profit = (Capacity – Throughput) * (P-V)

MC: Increased supply chain resilience may increase average costs


Reducing menu items might reduce customer choice, and reducing inventory


Fukushima, after they increased the buffer stock of critical components

Improve Supply Chain Resiliency

Increasing visibility into the lower tiers of a supply chain

Increases Average Flow Time

Throughput decreases, Higher throughput increases revenue

42% had Operations experience at some point in their career 31% were in Operations immediately before becoming CEO

Supply Chain Management Takeaways

  • Increased Supply chain resilience may increase average cost.
  • A firm’s supply chain should align with the business model.
  • New product development is an integral part of supply chain management.
  • A firm can have more than one tier of customers.
  • Decisions about suppliers affect customers.

Supply Chain Management

  1. Choosing suppliers
  2. Delivery
  3. Planning
  4. Production
  5. New product development
  6. Customer management

More New Balance

No significant moves with respect to 1.Level of domestic manufacturing 2. Acquisitions

2007 – Jim Davis steps down as CEO to make way for a new leader (Robert DeMartini) from outside the company (Tyson Foods)
By March 2008, NB had doubled its marketing budget to focus more on young consumers
DeMartini cut NB’s product line by 20% to refocus on best-selling products

Factors to Consider in Operations

  • Cost (production, logistics, use …) – Total Landed Cost, Total Cost of Ownership
  • Responsiveness (Lead Times for orders & new products, Delivery flexibility)
  • Quality (reliability, functionality, defects, …) – Technology (development, testing, …)
  • Support-ESG (carbon, pollution, labor conditions)
  • Relationship (prior experience, owner commitment)

Operations as a Strategic Differentiator

– Don’t try to beat others at what they do best (i.e., let Nike & Adidas/Reebok be the best marketers)
 Operations as a source of options – Domestic manufacturing → faster response capability → ability to serve retailers better & adjust to products with shorter life cycles