Operations Strategy¶

Course Overview¶

Operations Strategy is a 4-ECTS (European Credit Transfer System) course in the IESE MBA Year 1 curriculum, taught by Professor Mihalis G. Markakis (Department of Operations, Information and Technology). The course sits at the strategic layer above Operations Management: whereas OM addresses the tactical execution of a given operating system (inventory formulas, queueing, process flow), Operations Strategy addresses how to design that operating system and how to manage and change it over time to serve and enrich the firm's business strategy.

Central thesis: When designed coherently with business strategy, operations can not only deliver superior execution but also expand the set of feasible strategies a firm can pursue -- widening the competitive "battlefields." Operations is not merely the implementer of strategy; it is the foundation for, and driver behind, successful strategic attacks and defenses (Hayes & Upton, 1998).

Course type: Hybrid -- qualitative strategic reasoning (frameworks, case discussion) plus quantitative operations tools (critical fractile, inventory models, demand pooling).

Evaluation: Team presentations (50%), written assignment (10%), class participation (40%).

Session Map¶

Module 1 -- Capacity Strategy (Sessions 1-4)¶

Module 2 -- Process Strategy (Sessions 5-7)¶

Module 3 -- Supply Chain Management (Sessions 8-11)¶

Module 4 -- Developing Capabilities (Sessions 12-17)¶

Module 5 -- Transforming Operations (Sessions 18-21)¶

Core Frameworks¶

1. Operations-Based Strategy (Hayes & Upton, 1998)¶

Core argument: Strategic planning is not a top-down chess game where "grand plans" are formulated in the executive suite and then downloaded to operations. Rather, operations is the foundation behind successful strategic attacks and defenses. An operations-based competitive advantage is (a) embedded in people and processes, making it inherently difficult to imitate, and (b) less visible to competitors than a positioning-based advantage, so rivals are slow to respond.

Three types of operating capabilities:

Key examples from the reading: - Process-based: McDonald's meticulously documented procedures make its global consistency nearly impossible to replicate. Fidelity Investments' image and audio capture technology enables instantaneous, accurate customer service. - Systems-based: Allegheny Ludlum Steel coordinated small-batch customized steel production over six years, reducing defective output, doubling melt shop capacity (same equipment), and building a cost accounting system so precise it could estimate the cost of any grade/width/gauge of steel. Survived an industry consolidation (from 10 competitors to 3). - Organization-based: Lincoln Electric voluntarily shared its proprietary manufacturing methods with competitors during World War II. Competitors matched its costs by war's end, but Lincoln's organizational capability allowed it to rapidly regain its cost advantage. Boise Cascade brought a new paper plant to full capacity in one-third the time it took Union Camp using similar off-the-shelf technology.

Managerial implication: Companies should not just buy off-the-shelf technology; they must develop capabilities embedded in their people and operating processes so competitors cannot easily copy them. The higher up the capability hierarchy (from process to systems to organization), the more durable the advantage.

2. Ferdows' Six Strategic Roles of Foreign Factories¶

Framework dimensions: 1. Primary strategic reason for the site: Access to low-cost production, proximity to market, or access to skills/knowledge 2. Site competence: Scope of the factory's current activities, ranging from low (pure production execution) to high (innovation and global responsibility)

The six roles:

Upgrading path (three stages): 1. Improving the inside: Better layouts, employee training, cell production, Just-in-Time (JIT) manufacturing 2. Developing external resources: Developing local suppliers, managing global logistics, redesigning products for manufacturability 3. Taking on a global mandate: Becoming a center of excellence that generates new knowledge for the company's future

Operational warning: Upgrading requires a stable, robust global network and sustained corporate commitment. Continually threatening closure or shifting production based on short-term currency fluctuations undermines capability development.

3. Fisher's Supply Chain Framework¶

Core insight: The root cause of most supply chain failures is a mismatch between the nature of a product's demand and the design of its supply chain.

Two product types:

Two supply chain types:

The matching matrix:

Escape from the upper-right mismatch: Either (a) move down the matrix by investing to make the supply chain responsive, or (b) move left by stripping innovation from the product and competing as a functional commodity. For innovative products with 40% contribution margins and 25% stockout rates, the lost contribution from stockouts alone represents 10% of sales -- far exceeding the cost of building responsiveness.

4. The Critical Fractile (Newsvendor Model)¶

The critical fractile is the decision rule for single-period inventory problems -- situations where you get only one chance to order before uncertain demand is realized (seasonal fashion, perishable goods, newspapers, product launches).

Setup: - Price per unit: p - Cost per unit: c - Salvage value per unsold unit: v - Cost of underage (ordering too little): Cu = p - c - Cost of overage (ordering too much): Co = c - v

The critical fractile formula:

Critical Fractile = Cu / (Cu + Co)

This ratio determines the optimal service level. You should stock enough units so that:

P(D <= Q*) = Cu / (Cu + Co)

If demand is normally distributed with mean mu and standard deviation sigma:

Q* = mu + z* x sigma

where z* = NORM.S.INV(Critical Fractile) in Excel.

Interpretation: - High critical fractile (underage cost dominates) --> order aggressively (z is positive and large) - Low critical fractile (overage cost dominates) --> order conservatively (z is small or negative)

Example (from course materials): A book publisher sells at EUR 30, production cost EUR 9, no salvage value for the simplified case. Cu = 30 - 9 = EUR 21. Co = 9 - 0 = EUR 9. Critical fractile = 21/(21+9) = 0.70. With demand distributed as given, the optimal print run is 1,100 copies, yielding an expected profit of EUR 5,520.

Applications beyond inventory: Airline seat protection (how many seats to reserve for business class), capacity planning under demand uncertainty, hospital staffing, hotel yield management.

5. Inventory Management Systems (OITN-0011 through OITN-0013)¶

Continuous review system (Parts A and B): - How much to order: Economic Order Quantity (EOQ) minimizes total ordering + holding costs - When to order: Reorder Point = d x LT + Safety Stock - Safety Stock = z x sigma x sqrt(LT) - z reflects the desired service level (e.g., z=1.65 for 95%, z=2.33 for 99%)

Periodic review system (Part C): - Inventory reviewed at fixed intervals R; order placed to raise inventory position to base-stock level S - Vulnerable period: VP = LT + R (longer than continuous review) - Order-up-to level: S = d x (LT + R) + z x sigma x sqrt(LT + R) - Requires more safety stock than continuous review because uncertainty accumulates over the longer vulnerable period

Key comparison:

6. Demand Pooling and the Square Root Law¶

Concept: Aggregating demand across multiple locations reduces total variability because independent demand fluctuations partially cancel out.

The square root law: If a company consolidates inventory from n locations into one, total safety stock scales with sqrt(n) rather than n:

Safety stock (1 location) = z x sigma_total x sqrt(LT)
sigma_total = sigma_single x sqrt(n)  (for identical, independent locations)

Implication: Centralizing n identical warehouses into one reduces total safety stock by a factor of sqrt(n). Going from 4 warehouses to 1 cuts safety stock roughly in half. This is the quantitative basis for centralized distribution strategies (as explored in the Nike case).

Tradeoff: Centralization reduces inventory costs but increases transportation time/cost and may reduce responsiveness to local demand. The optimal network design balances these forces.

7. The Bullwhip Effect¶

Definition: The phenomenon whereby small fluctuations in end-consumer demand are amplified as orders propagate upstream through the supply chain, causing increasingly volatile swings at each tier (retailer --> wholesaler --> distributor --> manufacturer).

Experienced in: The Beer Game simulation (Sessions 8-9), where teams playing different supply chain roles routinely generate oscillations and inventory crises despite stable end-consumer demand.

Four primary causes: 1. Demand signal processing: Each tier forecasts from its own orders (not end-consumer data), adding noise at each step 2. Order batching: Ordering in large, infrequent batches creates lumpy demand signals upstream 3. Price fluctuations: Promotions and forward-buying cause demand surges unrelated to consumption 4. Rationing and shortage gaming: When supply is scarce, customers inflate orders; when supply normalizes, they cancel

Countermeasures: - Share point-of-sale (POS) data across all tiers (visibility) - Reduce order batching through smaller, more frequent orders - Stabilize pricing (Everyday Low Price / EDLP strategy) - Allocate supply based on past sales, not current orders (prevents gaming) - Vendor-Managed Inventory (VMI) programs - Reduce lead times to dampen amplification

8. Vertical Integration Framework (Make vs. Buy)¶

Core question: Should the firm perform an activity in-house or outsource it to a supplier?

Decision criteria:

Explored in: Holaluz case (Session 11) -- whether an energy company should in-source photovoltaic installation services or adopt a hybrid model. Costco case (Session 10) -- why Costco in-sources poultry processing and eyewear manufacturing despite being a retailer.

9. The Hard Discount Operations Model (Aldi/Lidl)¶

Origin: Theo Albrecht, Germany, 1962. Born out of necessity: limited capital, customers with low disposable income, competing against established retail cooperatives.

Four pillars: 1. Limited assortment: Narrow range of categories, shallow depth within each category (~700-900 stock keeping units (SKUs) in-store vs. 100,000+ at traditional retailers). Reduces complexity costs and enables massive economies of scale per SKU. 2. Private label: 60%+ private label products. Gives discounter control over pricing and leverage with suppliers (can switch suppliers without losing the brand). 3. Quality: Stringent quality requirements. Suppliers participate in annual public quality tests. High quality at low price generates customer loyalty, compensating for the absence of brand recognition. 4. Efficient operations: Cross-docking distribution (goods flow through distribution centers (DCs) within hours, not days), hub-and-spoke logistics with DCs serving 50-80 stores within a 50 km radius, ~50 inventory turns per year (vs. ~7 at Walmart).

Operational efficiency details: - Purchasing power: EUR 40 billion sales / ~10,000 SKUs = EUR 4 million average sales per SKU (vs. trivial per-SKU volume at retailers with 100,000+ SKUs) - Store operations: Multi-functional staff, pallets placed directly on sales floor (no unpacking/shelving), standardized store layout unchanged since the 1970s, short checkout counters with no buffer zone to speed throughput - Logistics: Own trucking fleet, aerodynamic truck designs, retreaded tires, 4-6 deliveries per week to each store - Cellular growth: When a cell reaches 80 stores, a second DC is built, the cell divides, and growth continues -- financed entirely from operating cash flow

Negative working capital model: - Average collection period: ~7 days (products sold within a week) - Payment to suppliers: 30 days - Gap: 23 days of negative working capital - At EUR 150 million daily sales, this creates a EUR 3.45 billion constant cash balance

Key insight: The hard discount model is a prime example of strategic coherence where every element reinforces another. The limited assortment enables economies of scale, which makes the company attractive to suppliers. The narrow range of everyday products sold through 50-80 stores creates stable demand, which enables cross-docking. Fast inventory turns enable negative working capital, which funds growth without external financing.

Retail format map (two dimensions): - X-axis: Number of categories (breadth) - Y-axis: Number of SKUs per category (depth) - Hard discounters occupy the "sweet spot" of operational efficiency in the bottom-left corner - Moving away from that corner increases operational expenses (more logistics modes, slower-turning items, greater complexity)

10. The Fast Fashion Operations Model (Zara/Inditex)¶

Value proposition: Fashionable clothing at accessible price points (Value = Benefit - Cost). Competes on both product freshness and price.

Two operational pillars:

Pillar 1 -- Quick Response (QR) production: - Postpone all risky production decisions until evidence of market demand exists - Reduces traditional design-to-store timeline from ~21 months to ~4 months - Near-shore suppliers (Spain, Portugal, Morocco for European markets) enable lead times of weeks instead of months - Item-by-item production (not seasonal collections) smooths resource utilization year-round - Key metric: Gross Margin Return on Inventory (GMROI) = Gross margin / Average inventory

Pillar 2 -- Dynamic assortment planning: - Store offerings updated weekly, every third day, or even daily (vs. twice per year under traditional model) - New product introductions timed using optimization models (similar to film release scheduling) - Closed-loop control: real-time sales data feeds back into assortment decisions - Research shows revenues can increase by up to 10% using dynamic assortment planning - Rule of thumb: basic products introduced at season start; fashionable items spaced throughout the season to refresh the assortment

Design process differences: - Traditional: design starts 14 months before sale, full collection presented to wholesale buyers - Fast fashion: works at item level, eschews wholesale channels, designs respond to nascent demand trends, raw materials pre-positioned for rapid order fulfillment

Distribution differences: - Traditional: large initial store loading + replenishment + end-of-season markdowns - Fast fashion: last-minute distribution decisions send inventory where most needed, smaller more frequent shipments

Sustainability concerns: waste (clothing purchases per capita doubled between 2000-2012), working conditions (Rana Plaza collapse 2013, Accord on Fire and Building Safety in Bangladesh), offshoring destroying local supplier ecosystems (Italian denim suppliers, Spanish QR suppliers). Inditex's "Working in Clusters" program addresses some concerns across 10 countries covering 85%+ of production.

Applicability beyond fashion: The model applies wherever products rotate frequently and consumers seek novelty -- food (Seven-Eleven Japan), consumer electronics, entertainment (film release timing, museum exhibits).

11. AI in Operations¶

Explored in: Session 7 (Arriaga Asociados case -- AI as paralegal tool in a law firm).

Key application areas: - Demand sensing: Machine learning (ML) models that detect demand shifts in real-time from POS data, social media, weather, and events -- improving on traditional statistical forecasting - Predictive maintenance: Sensor data and ML algorithms predict equipment failures before they occur, converting unplanned downtime into planned maintenance windows - Process automation: Robotic Process Automation (RPA) and AI-driven decision systems that handle routine operational decisions (order routing, scheduling, quality inspection) - Supply chain optimization: AI-powered tools for route optimization, inventory positioning, supplier risk assessment

Strategic tradeoffs (from the case): - How does AI adoption affect the firm's operations vs. its business model and strategy? - When does AI augment human workers vs. replace them? - Implications for organizational capabilities and change management

12. Technology Adoption and Change Management¶

Change management (Session 18 -- Imperia SCM case): - The SaaS vs. consulting firm dilemma: technology companies must decide whether their primary value comes from software (scalable) or from bundled services (high-touch but less scalable) - Onboarding and change management programs are critical for clients to extract value from operational software - The Chief Operating Officer (COO) role in driving company-wide technology adoption

Business Process Reengineering (BPR) (Session 19 -- Porsche 1992 case): - One of the best-executed turnarounds in business history - Analysis of market positioning, cost structure, product pipeline, operations, and supply chain in crisis - Role of lean manufacturing principles in the turnaround - Short-term stabilization vs. medium-term operational transformation vs. long-term strategic repositioning

Business model innovation (Session 20 -- Netflix case): - Sequential technology disruptions (VHS --> DVD/early internet --> broadband/streaming) - Each business model requires different processes and capabilities - Netflix's repeated self-transformation demonstrates organization-based capabilities (Hayes & Upton framework)

13. Supply Chain Resilience (Triple-A Supply Chain)¶

Concept from Hau Lee (Harvard Business Review, 2004): The best supply chains are not merely efficient; they possess three qualities:

1. Agility: Ability to respond quickly to short-term demand/supply changes
2. Adaptability: Ability to adjust the supply chain design to structural market shifts
3. Alignment: Ability to create incentives so all supply chain partners' interests are aligned

COVID-19 disruption lessons (Session 13): - Packed ports and ships coexisting with empty shelves exposed fragility in globally interconnected supply chains - Lean practices (minimal inventory) amplified disruption effects - Winners and losers were determined by supply chain resilience, not just efficiency - Raised fundamental questions about whether globalization and interconnectedness make supply chains more resilient or more fragile

Quantitative Tools -- Back-of-Napkin Reference¶

Capacity Sizing Heuristics¶

• Critical fractile for capacity/inventory: Fc = Cu / (Cu + Co). Compute, read off the cumulative distribution function (CDF) of demand, find optimal quantity.
• Break-even utilization: Minimum capacity utilization to cover fixed costs = Fixed costs / (Price - Variable cost per unit) / Maximum capacity
• Demand pooling rule: Consolidating n independent, identical demand streams into one reduces safety stock by a factor of sqrt(n)

Offshoring Cost Comparison¶

When comparing onshore vs. offshore production, include: - Unit labor cost difference (the obvious factor) - Transportation and logistics costs (container shipping, lead time) - Inventory carrying costs (longer lead times require more safety stock: SS = z x sigma x sqrt(LT)) - Quality costs (defect rates, rework, inspection) - Coordination costs (time zones, language, travel) - Intellectual property (IP) risk - Flexibility penalty (inability to respond to demand changes during long lead times -- Fisher's framework) - Currency risk and tariff exposure

Rule of thumb: If the offshore landed cost (unit cost + shipping + inventory + quality + coordination) exceeds ~85% of the onshore cost, the intangible risks and loss of responsiveness often make onshoring the better choice.

Supply Chain Responsiveness vs. Efficiency Tradeoff¶

Fisher's framework quantified: - For an innovative product with 40% contribution margin and 25% stockout rate, lost contribution = 0.40 x 0.25 = 10% of sales - Every dollar invested in responsiveness that reduces stockouts yields more than a dollar in recovered margin - For a functional product with 10% margin and 1% stockout rate, lost contribution = 0.10 x 0.01 = 0.1% of sales -- negligible, so invest in efficiency instead

Inventory Formulas Quick Reference¶

Quick Reference¶

1. Operations-Based Strategy (Hayes & Upton): Operations is the foundation for strategic attack and defense, not just the implementer of strategy. Three capability types: process-based (moderate to imitate), systems/coordination-based (hard to imitate), organization-based (hardest to imitate). → See: Operations Strategy
2. Ferdows' Six Factory Roles: Offshore, Source, Server, Contributor, Outpost, Lead. Categorized by strategic reason (cost, market, knowledge) and site competence (low to high). Upgrade path: improve inside → develop external resources → take global mandate. → See: Operations Strategy
3. Fisher's Supply Chain Framework: Functional products need efficient supply chains; innovative products need responsive supply chains. The most common failure is pairing innovative products with efficient supply chains. → See: Operations Strategy
4. Critical Fractile (Newsvendor): Fc = Cu / (Cu + Co). Optimal order Q = mu + z x sigma where z* = NORM.S.INV(Fc). High underage cost → order aggressively. → See: Operations Strategy
5. EOQ: Q* = sqrt(2DS/H). Balances ordering cost against holding cost. → See: Operations Strategy
6. Reorder Point: ROP = d x LT + z x sigma x sqrt(LT). Safety stock covers demand variability during lead time. → See: Operations Strategy
7. Demand Pooling / Square Root Law: Consolidating n identical locations into one reduces total safety stock by a factor of sqrt(n). Tradeoff: lower inventory cost vs. higher transport time. → See: Operations Strategy
8. Bullwhip Effect: Small demand fluctuations amplify upstream. Four causes: demand signal processing, order batching, price fluctuations, shortage gaming. Fix with POS data sharing, smaller orders, stable pricing, VMI. → See: Operations Strategy
9. Hard Discount Model (Aldi/Lidl): Four pillars: limited assortment, private label, quality, efficient operations. Negative working capital (sell in 7 days, pay suppliers in 30) funds growth. → See: Operations Strategy
10. Fast Fashion (Zara): Quick Response production (postpone risky decisions until demand evidence exists) + dynamic assortment planning (refresh store offerings weekly). Key metric: GMROI = Gross margin / Average inventory. → See: Operations Strategy
11. Triple-A Supply Chain (Hau Lee): Agility (respond to short-term changes), Adaptability (adjust design to structural shifts), Alignment (align partner incentives). Efficiency alone is fragile. → See: Operations Strategy
12. Vertical Integration (Make vs. Buy): Favor make when: core competency, high asset specificity, few suppliers, tight coordination needed. Favor buy when: non-core, competitive supply base, rapidly changing technology. → See: Operations Strategy
13. Continuous vs. Periodic Review: Continuous = fixed order quantity at variable timing (safety stock covers LT). Periodic = variable quantity at fixed intervals (safety stock covers LT + R, so requires more). → See: Operations Strategy
14. Business Process Reengineering (Porsche): Short-term stabilization, medium-term operational transformation via lean principles, long-term strategic repositioning. → See: Operations Strategy

Cross-References¶

• Operations Management: Operations Strategy is the strategic layer on top of OM's tactical tools. OM provides the execution machinery (EOQ, queueing, process analysis); Operations Strategy determines which operating system to design and how to evolve it. The OITN technical notes on inventory management (Parts A-D) bridge both courses.

• Competitive Strategy: Operations as a source of competitive advantage. Hayes & Upton's operations-based strategy complements Porter's positioning view -- sustainable advantage can come from embedded operational capabilities (process, systems, organization), not just market positioning. Fisher's framework connects product strategy (functional vs. innovative) to supply chain design.

• Corporate Finance: Capacity investment decisions are capital budgeting problems (Net Present Value (NPV) analysis of building new plants, expanding capacity, or investing in supply chain responsiveness). The hard discount model's negative working capital is a financial strategy enabled by operational efficiency. Inventory is working capital that ties up cash.

• Marketing: Demand forecasting is the input to all supply chain decisions. Fisher's framework starts with classifying product demand patterns. The bullwhip effect demonstrates how poor demand signal processing destroys supply chain performance. Dynamic assortment planning (fast fashion) connects product merchandising to operational execution.

• Global Economics: Offshoring decisions involve labor cost arbitrage, exchange rate risk, tariff policy (Mercosur, NAFTA), and trade agreements. Ferdows' framework explicitly addresses why firms manufacture abroad and how geopolitical changes force network reconfiguration.

• Analysis of Business Problems / Decision Analysis: The critical fractile is a direct application of expected value (EV) decision-making under uncertainty, connecting to decision trees and probability distributions taught in those courses.

Key Readings¶

Textbooks (Reference)¶

• Moscoso, P. and Lago, A. (2017). Operations Management for Executives. McGraw-Hill.
• Slack, N. and Lewis, M. (2024). Operations Strategy, 7th Edition. Pearson.
• Chopra, S. (2019). Supply Chain Management, 7th Edition. Pearson.

Glossary¶