!! ! ! ! !! ! ! ! !! ! ! !! !! !! !! !! !! ! ! ! !! ! ! !! !! ! ! ! !! ! ! ! ! !! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !

Introduction to OM Diff between products and services: • Tangibility, measurement of productivity, quality evaluation, level of customer involvement, definition consistency, ability to inventory, patentability Competition in OM • Product and service design, process and capability, quality, inventory and materials, quick response and flexibility, supply chain management, sustainability Productivity: Performance measurement of the effective use of resources ∝ 1/cost Goods/Services produced Inputs used to produce them

Flow Time (FT): Time taken for unit to flow through entire process = 1/Inventory Turns Flow Rate (FR): Throughput/output rate • Constrained by demand rate, process capacity and supply rate Inventory: Average no of flow units in the process Little’s Law • Condition: The system is in steady state

Process Design/Selection • Decision on the way production of goods and services will be organised and transformed

!

Trade-off: An exchange where you give up one thing in order to get something else that you also desire. Operations function: Part of an organisation that is responsible for its processes that design, manage, improve and deliver its products/services to its customers Role of OM • Improve productivity, matching supply with demand, improving efficiency, manage trade-offs
Product Design, Process Design, Capacity Planning

! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !

Process Flexibility: Degree to which system can be adjusted to changes in processing requirements due to factors such as product and service design changes, product volume changes, changes in process technology

Inventory = FR x FT

Utilisation • Measures what resource/process actually produce compared to what it can produce at full speed Resource/Process Utilisation = Flow rate Resource/Process Capacity

Developing Capacity Alternatives • Design flexibility, account for stage of life cycle, attempt to smooth capacity requirements, prepare to deal with capacity “chunks”, take a “big picture” approach to capacity changes, identify optimal operating level (volume vs costs), choose a strategy if expansion is involved (leading, following, tracking) • Eg. Capacity cushion: Extra capacity built to offset demand uncertainty = 100% - process utilisation Factors to consider when outsourcing • Available capacity, expertise, quality considerations, nature of demand, costs, risks Capacity planning in Service • Challenges: Need to be near customers, inability to store services for consumption later, degree of demand volatility is higher • Strategies to offset capacity limitations to achieve closer match between supply and demand: Pricing, promotions, discounts, tactics to shift demand from peak periods to slow ones
Process Selection, Facility Layout, Line Balancing

Driving forces for (re)design • Competitiveness, cost/availability, technology (i.e.patent), economic growth, social and demographic, political, liability, legal, sustainability Key Design Issues • Product/service life cycle, standardisation (lack of variety), mass customisation (modular design, delayed differentiation), product/service durability, reliability, aesthetics etc, degree of newness, global design, sustainability Design Focus • Tuning Scaling Support Transition

Challenges in Service Design • Barriers are low to enter and exit, focus on intangible factors, service design and delivery process cannot be separated, outputs vary, customers are part of inputs and processes, cannot be inventoried (flexibility and capacity), importance of location Design for Operations • Account for the capabilities of the organisation’s operations function • Eg. Concurrent Engineering (bringing people from different sectors like purchasing and marketing together early in the design stage) • 3DCE (3-Dimensional Concurrent Engineering): Simultaneous development of product, process and supply chain • Other key terms: Computer-aided design (CAD), Design for manufacturing (DFM), Design for Assembly (DFA), Design for Disassembly (DFD), Design for Recycling (DFR) Trade-off in Design • Responsiveness vs Customisation

Basic Layout Objectives • Facilitate smooth flow of work, material and information through the process by utilising resources efficiently and effectively Process Layout • Not arranged according to a particular production sequence • Can handle varied processing requirements • Processes jobs of same requirements in batches
Activity time = Setup time + Run time

Implication: Resource Capacity = Batch size (Q) Setup time + (Q x Activity time)

!

! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !

High variety Batch Production

High volume Repetitive Production Line Cycle time, speed

Job Shop Setup time, high flexibility

! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !

When m ≠ 1, Utilisation of each server(ρ) = λ / mμ Average customers in service = mxρ Average customers in process(Ls) = Ws x λ = Lq + m x ρ

! !

Difference between variability and seasonality • Variability is inherent and may be addressed by holding excess capacity; seasonality is addressed by staffing to meet expected demand Flow rate = Arrival rate Process Capacity = Service rate Quality Management Design Quality • Embedded in the design specifications of the product/service Conformance Quality • Depends on production and delivery processes Random variation due to common causes • Natural variation in the output of a process created by countless minor factors Assignable variation due to special causes • Variation whose cause can be identified; can be beneficial Statistical Process Control (SPC) • Statistical evaluation of the outputs of a process to detect whether there has been a change in the underlying steps; nothing to do with making good/bad products • In general: 1. Identify whether process variation are random/assignable 2. Identify reasons behind variation 3. Reduce process variability 4. Take corrective action/Institutionalise positive changes if necessary Sample statistics are used to judge the • randomness of the process variation • Distribution of sample statistics ~ N(μ, σ2)

! ! ! ! ! ! ! ! ! ! ! ! !

If process mean is centred between USL and LSL, Cp = USL - LSL 6σ

If process mean is not centred, Cpk = min{ USL - x x - LSL ̅ ̅ , } 3σ 3σ

Improve process capability • Simplify, standardise, mistake-proof, upgrade equipment, automate ISO 9000 • Implemented broadly to certify a firm’s quality management process • Monitored by International Standards Organisation (ISO) • Ensures consistency in processes, but not necessarily quality of products QM Tools • The 5 whys, pareto analysis, cause-and-effect diagram, 6σ quality (ideal: have no more than 1.98 defects ppb; real: have no more than 3.4 defects ppm) Total Quality Management (TQM): Philosophy that involves everyone in an organisation in a continual effort to improve quality and achieve customer satisfaction Inventory Management I: ABC Analysis & EOQ Inventory • Types: Work-in-progress (WIP), raw materials, finished goods, maintenance, repairs and operating (MRO), goods-in-transit (pipeline inventory) • Functions: Manage production (permit operations, smooth production requirements, decouple operations), meet demand (meet anticipated demand, protect against stock outs), control costs (take advantage of order/ production cycles, hedge against price increases, take advantage of quantity discounts) • Reasons not to hold inventory: Expenses, obsolescence, delayed responsiveness, masking underlying problem

Management of Queuing & Waiting Why does queue form? • Variability in inter-arrival/service time and high utilisation Flow time(Ws) = Waiting time(Wq) + Service time(1/μ) Inventory = Queue length + Customers in service Flow rate = Customer arrival rate

! !

! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !

Minimum TIC,

! ! ! !

TIC =

Q D H+ S 2 Q

Q* =

then, TIC =

Re-Order Point (ROP) • Inventory level which new order should be placed to avoid future shortage If LT ≤ CT, ROP = LT x D Rate = (LT/CT) x Q

If LT ≥ CT, ROP = fraction[LT/CT] x Q

EOQ with Quantity Discount & EPQ Robustness of EOQ • Cost-robust because 100% increase/50% decrease in EOQ only result in 25% error in TIC • TIC curve is flatter to the right • EOQ increase with demand rate and fixed ordering cost and decreases with unit carrying costs; unit purchasing costs do not affect EOQ TC = TIC + P x D

Quantity Discount • Q* can be a feasible EOQ or a discounted Q When H is Fixed • Compute EOQ normally, only one price curve will have the EOQ in its feasible price range (this curve represents upper bound on what you will pay per unit), compute TC for ordering that EOQ and TC at the price break for the subsequent lower price curves When H is not Fixed • Begin with the lowest unit price, compute the EOQ for each price range until you find a feasible EOQ; if EOQ for lowest unit price is feasible, it is the optimal order quantity; if EOQ is not feasible in the lowest price range, compare TC at the price break for all lower prices with the TC of the feasible EOQ Economic Production Quantity (EPQ) • Assumptions: Only 1 product is produced, annual demand known, constant and continuous usage rate, periodical production, production rate constant, produce in batch of Q, setup time does not vary, there are no quantity discounts

Assumptions • System in steady state: Service rate>Arrival rate • First Come, First Serve (FCFS) • Arrivals independent of one another

Where CV1/λ = λ x σ1/λ; CV1/μ = μ x σ1/μ When m = 1, Average customers in queue/queue length(Lq) = Wq x λ Average customers in service(ρ) = λ / μ = Process utilisation

! ! ! !

!

Average customers in process(Ls) = Ws x λ = Lq + ρ

! !

Voice of the process • Time ordered plot of representative sample statistics obtained from an ongoing process that can be used to distinguish between random and non-random variability Control Charts • Components: Center line for expected performance, UCL (μ + 3σ) and LCL (μ - 3σ) • Purpose: Monitor process outputs to see if it is in control • 2 types: Mean (x-bar) and Variation (Range) • 2 outcomes: “In control” (all data fall randomly within UCL and LCL) and “out of control” Possible signs of “Out of Control” • 1 point near/outside 3σ control limits, run of 7 points in a row, cycles or other non-random patterns ,2 out of 3 points in a row outside of either 2σ limits, 4 or 5 points in a row outside of either 1σ limits • Type I error (manufacturer’s risk), type II error (consumer’s risk) Process Capability (Process is “in control”) • Determine if process is capable of producing output that is within specifications established by engineering design or customer requirements • Condition: Cp or Cpk ≥ 1.33 (i.e. 8σ range)

Efficient & Effective Inventory Management • Reliable forecast of demand, knowledge of mean lead time and its variability, reasonable estimates of excess costs per unit inventory/per unit inventory shortage costs/fixed inventory management costs, inventory counting system (periodic system, perpetual system, universal product code (UPC), radio frequency identification tags (RFID)), classification for inventory items (ABC analysis), inventory management models (EOQ & quantity discount, EPQ, newsboy)

! !

Product Layout • Arranged according to a particular production sequence • Use standardised processing operation to achieve smooth, rapid, high-volume flow Cycle Time: Flow time at each station = 1/ process capacity Line Balancing: Process of assigning tasks to workstations in such a way that the workstations have approximately equal time requirements • Use labour and equipment more efficiently • Avoid fairness issue • Done by: Re-allocating tasks, replicating the line, selectively adding workers, further specialisation

Measures to reduce waiting • Reduce variability in arrival process (schedule appointments, peak-load pricing, renewals by mail, online boarding passes) and in service process (improve processing rates, priority rules like FCFS and SPT, self-service) • Reduce utilisation by increasing mean service rate (scheduling) or reducing mean arrival rate • Increasing capacity/service rate by adding servers To determine minimum number of servers •ρ