Supply Chain Management

Definition

A supply chain encompasses all parties involved — directly or indirectly — in fulfilling a customer request. This includes manufacturers, suppliers, transporters, warehouses, retailers, and ultimately the customer.

Supply chains manage three key flows simultaneously:

• Product flow: Materials and goods moving downstream from suppliers to customers.
• Information flow: Demand signals, orders, forecasts moving upstream from customers to suppliers.
• Financial flow: Payments, credit terms, invoicing across the chain.

Cycle View

Supply chain processes are divided into cycles, each at the interface between two consecutive stages. The four standard cycles are:

Push / Pull View

The push/pull boundary divides speculative activity from reactive activity:

• Pull processes are initiated by an actual customer order — reactive, Build-to-Order. Example: Dell building a laptop to a specific customer configuration.
• Push processes are executed in anticipation of future demand — speculative, Make-to-Stock. Example: Zara producing a collection before knowing actual store sales.

Every firm in a supply chain participates in three macro-level process groups that must be tightly integrated:

The SCOR model extends this into six operational pillars: Plan, Source, Make, Deliver, Return, Enable — providing a cross-industry standard language for supply chain benchmarking and management.

Supply chain decisions operate at three distinct time horizons:

Marshall Fisher (Harvard Business Review, 1997) provides the foundational framework for matching supply chain design to product characteristics. The key insight: not all products need the same supply chain.

Two Product Types

Two Supply Chain Types

Strategic fit is the alignment between a company's competitive strategy and its supply chain strategy. Both must target the same goal — one consistent with customer needs.

• Competitive strategy: What customer needs does the company serve? (price, speed, variety, reliability, quality)
• Supply chain strategy: What capabilities does the supply chain build? (efficiency, flexibility, responsiveness)

Demand uncertainty is shaped by: range of quantity required, lead time demanded, variety of products needed, required service level, rate of innovation, and acceptable price point.

Implied demand uncertainty is not raw demand variability — it is the uncertainty your supply chain must absorb, after accounting for your product type and customer service commitments.

Factors that increase implied uncertainty:

• A wider range of quantities required (high/low demand peaks)
• Shorter lead times demanded by customers
• Greater product variety and SKU proliferation
• Higher required service levels (fill rates, availability)
• High rate of product innovation (short life cycles)

The Strategic Challenge

Apple competes on premium innovation — highly innovative products with uncertain demand and high margins. This demands a responsive supply chain. Yet Apple also needs to produce at extraordinary scale (hundreds of millions of units), which demands elements of efficiency. The case shows how Apple balances this tension.

Key Strategic Moves

• Foxconn partnership (2001/2005): Apple offshored assembly to China, leveraging scale and skilled labour. "iPhone City" in Zhengzhou cost $10B+ and could produce 500K+ iPhones/day.
• Vertical integration in chips: Acquiring P.A. Semi in 2008 allowed Apple to design its own A-series processors (A4 chip, 2010), reducing dependence on Samsung and Qualcomm.
• TSMC partnership: Taiwan Semiconductor supplies Apple's chips — a critical, highly managed supplier relationship.
• Supplier funding: Apple pre-pays or part-funds supplier equipment, securing capacity and locking in supply — converting suppliers into dedicated partners.

Geopolitical Risk — The China Dependency

Between 2015 and 2019, Apple's China supplier share grew from 44% to 47%. This concentration poses significant risk. Post-COVID, Apple began diversifying toward India (8 new supplier locations) — a strategic supply chain adaptation in response to changing geopolitical realities.

Scope of Strategic Fit

Strategic fit cannot be achieved at the product level alone. As product portfolios grow and customer segments diversify, the scope expands:

• Intra-company fit: All functions (marketing, operations, finance) must support the same SC strategy.
• Inter-company fit: All supply chain stages (suppliers, manufacturers, distributors, retailers) must be aligned.
• Tailored supply chains: Different product lines or customer segments may require different supply chains from the same firm (e.g., Apple's direct online channel vs. retail partners).

Common Exam Traps

The bullwhip effect describes the phenomenon where small fluctuations in customer demand become increasingly amplified as orders move upstream through the supply chain — from retailer to wholesaler to distributor to manufacturer.

The consequences for manufacturers include: factories running overtime at some periods and idle at others, huge safety stock requirements, premium freight costs, excess markdowns, and reduced customer service levels — all generated by demand variability that does not reflect actual consumer behaviour.

Lee, Padmanabhan & Whang (1997) identified four root causes. Each is a consequence of rational decision-making within an imperfect information structure:

1. Demand Forecast Updating

Each upstream stage uses the orders it receives as its primary demand signal, then adds safety stock to cover uncertainty and lead time. Each layer of exponential smoothing amplifies the variance. With longer lead times, safety stock requirements explode.

2. Order Batching

To reduce fixed ordering costs and fill truckloads efficiently, companies order in large batches periodically (e.g., once a week or month). This creates large spikes followed by zero orders, which upstream stages cannot distinguish from true demand swings.

3. Price Fluctuation (Trade Promotions)

When manufacturers offer periodic promotions or quantity discounts, buyers rationally forward-buy — stocking up far beyond current needs. This creates artificial demand peaks followed by deep valleys. P&G found this was "the dumbest marketing ploy ever."

4. Rationing and Shortage Gaming

When supply falls short, manufacturers often allocate in proportion to orders placed. Knowing this, buyers over-order to ensure they receive enough. When the shortage ends, orders vanish instantly. HP's LaserJet III rationing caused phantom orders and millions in excess inventory.

Lee et al. (1997) organise mitigation strategies around three mechanisms:

Company Background

Barilla SpA, the world's largest pasta producer, operated 800+ dry product SKUs across a complex Italian distribution network: two Central Distribution Centers (CDCs) supplying Grand Distributors (GDs) and Organised Distributors (DOs), who in turn supplied 100,000 retail outlets.

The Problem

Weekly orders from the Cortese Northeast DC showed extreme variability: mean of 300 quintals, standard deviation of 227 quintals — a coefficient of variation of 76%. Actual consumer demand was essentially flat. The variability was entirely generated within the supply chain through promotions, batching, and information distortion.

Brando Vitali's JITD Proposal

Just-in-Time Distribution: instead of delivering based on whatever distributors order, Barilla would determine the "appropriate" delivery quantities itself — based on downstream data — to better meet consumer needs and smooth Barilla's manufacturing load. Essentially VMI before VMI had a name.

Why It Failed (Initially)

Significant resistance on both sides:

• Distributors: "Managing stock is my job." Fear of losing autonomy. Concern about giving Barilla power to push product into warehouses.
• Barilla's own sales team: Feared losing promotional flexibility. Worried about losing distributor shelf space if inventory decreased. Doubted cost savings.

Seven-Eleven Japan (SEJ) was founded in 1974 and became the world's largest convenience retail chain, with over 21,000 stores in Japan by 2021. Its core model is built around high-frequency, small-lot replenishment driven entirely by real-time demand data — the opposite of bulk, forecast-driven restocking.

The business proposition: convenience over price. SEJ stores are intentionally small (close-by) and stocked based on hyper-local demand patterns — age, gender, time-of-day, weather. When Japan's ageing population grew, SEJ added the Seven-Meal home-delivery service and expanded ready-meal lines accordingly.

New products are trialed for just 3 weeks: if a product does not contribute to sales and margin within that window, it is delisted. This rapid SKU rotation requires a supply chain that can onboard, track, and exit products at speed.

The fundamental supply chain architecture question for multi-site retailers: should each store manage its own orders, or should a central system (or DC) control replenishment?

SEJ operates separate supply chains by temperature: frozen (3–7x/week), chilled and warm foods (3x/day), room-temperature (1x/day). This granularity is only possible because centralized data allows precise planning across the whole network.

SEJ's Total Information System, launched in 1979 and upgraded continuously, is the operational backbone. It links every store with HQ, suppliers, and distribution centres in real time.

Store hardware includes a graphic order terminal (handheld, used by store manager to place orders using POS analytics), a POS register (captures age/gender of buyer), a store computer, and a scanner terminal (reconciles deliveries without truck driver waiting). Each piece eliminates friction and compresses the order-to-shelf cycle.

The 1995 Kobe earthquake is the canonical example of SEJ's agility. When the earthquake struck, road infrastructure collapsed and normal distribution routes became impassable.

This was possible because of three pre-existing capabilities:

• Real-time demand visibility — HQ could see exactly which stores were operational and what they needed
• Supplier relationships and trust — SEJ's aligned partner network could be activated without lengthy contract renegotiation
• Backup logistics planning — alternative transport modes (helicopter, motorcycle) were pre-arranged, not improvised

This episode is directly cited in Hau Lee's Triple-A Supply Chain article as the benchmark example of Agility in action — the ability to respond to sudden, short-term disruptions without losing momentum.

SEJ is the case study that best illustrates all three pillars of Lee's Triple-A framework simultaneously:

The fundamental operating logic of any supply chain is determined by where the push/pull boundary sits relative to the customer order point.

The Newsvendor Model solves a classic single-period inventory problem: how much to order when demand is uncertain and unsold units cannot be carried over (perishables, fashion, newspapers).

Where: p = selling price, c = unit cost, s = salvage value.

Obermeyer application: At a wholesale price of $112.50, profit margin is ~24% ($27/unit) and liquidation loss is ~8% ($9/unit). So Cu = $27, Co = $9, CR = 27/(27+9) = 0.75 — meaning Obermeyer should produce enough to satisfy 75% of demand scenarios, erring on the side of availability.

When demand uncertainty is high (wide bell curve), the stakes of the Newsvendor decision are much larger — small errors in Q* lead to large cost penalties. This is exactly Obermeyer's challenge: styles like Anita (SD = 1,047) vs. Gail (SD = 194) have completely different risk profiles requiring different production strategies.

Sport Obermeyer is a premium skiwear manufacturer based in Aspen, Colorado, facing the classic "fashion gamble": production decisions must be made 12–18 months before the selling season with minimal market information.

The Buying Committee model: Six managers independently forecast demand for each style. The standard deviation of these forecasts, multiplied by 2, estimates actual market demand variance. High committee disagreement = high market risk = defer to reactive phase.

Two-phase production logic:

• Phase I (Speculative, Nov–Feb): Commit 10,000 units to high-consensus, low-variance styles. Use China (low cost, high MOQ) for these stable volumes.
• Phase II (Reactive, post-Las Vegas show): Use early order data (first 20% of season orders) to update forecasts dramatically. Reserve Hong Kong capacity (low MOQ = 600 units, 1–2% repair rate) for late, precise runs on the "hits."

Risk pooling is the statistical principle that aggregating independent demand streams reduces relative variability. It is one of the most powerful tools in supply chain design.

This is why centralised inventory outperforms decentralised inventory for uncertain products. By pooling demand across locations, the total safety stock needed falls — even while total demand stays the same.

Product redesign as risk pooling: Obermeyer reduced zipper variety fivefold by using black zippers as a design element across multiple lines. This meant one common component (pooled) serves many end products — reducing both procurement risk and inventory cost.

Zappos was founded in 1999 with one radical idea: sell shoes online. By 2008 it was approaching $1 billion in gross sales — not by being the cheapest, but by delivering a service experience so exceptional it became a competitive moat. The phrase "Deliver WOW Through Service" is Zappos Core Value #1.

The supply chain evolution:

Random stocking: Items are placed in any available bin (not by brand or category). This eliminates picking bottlenecks — workers always find nearby inventory — and ensures 100% accuracy: if an item shows online, it is physically available. When the last unit of a style/colour/size is sold, it disappears from the website immediately.

Call centre as competitive weapon: Phone number prominent on every web page. Calls answered in under 20 seconds. Operators empowered to direct customers to competitors if Zappos is out of stock. This is alignment in action — the supply chain serves the customer relationship, not the other way around.

Reverse logistics covers all flows of products moving backwards through the supply chain — from customer back to manufacturer, refurbisher, or recycler. In e-commerce, return rates of 20–35% are common; ignoring this flow is not an option.

Why reverse logistics is hard:

• Returns are difficult to forecast — timing, volume, and condition are all uncertain
• Flows are less visible in accounting systems — costs are diffuse and hard to attribute
• Reverse flows are low volume, high variety — the opposite of what forward supply chains are optimised for
• Requires separate processes and facilities — cannot simply run the forward chain in reverse

Amazon's easy returns policy — free returns at 18,000 drop-off locations, no box or label required at Kohl's and UPS — set a new industry baseline. But the resulting flood of returned goods created an enormous operational and sustainability challenge.

The returns problem: Easy returns drive purchase volume (good) but create massive reverse logistics complexity (costly). Returned items must be inspected, sorted, re-stocked, liquidated, or destroyed — each at different costs. Amazon faced a sustainability crisis: millions of returned items were reportedly being landfilled because processing costs exceeded resale value.

Amazon's strategic responses:

• Returnless refunds — for low-value items, Amazon issues a refund without requesting the item back. Processing cost exceeds item value.
• Grading and resale channels — returned goods graded and sold via Amazon Warehouse (like-new) or Amazon Liquidations (bulk lots to resellers)
• Recommerce partnerships — working with third parties to refurbish and resell electronics and apparel
• Reducing returns at source — better product photos, fit prediction tools (AR try-on), detailed sizing guides to reduce fit-related returns

Industry effect: Amazon's standard forced Walmart, Target, and traditional retailers to match easy-return policies — raising the entire industry's reverse logistics cost base. This is an example of how one supply chain decision by a market leader reshapes competitive dynamics industry-wide.

Session 6 brings together two themes that are increasingly inseparable: customer-centricity (designing the supply chain around the customer experience) and sustainability (designing it around long-term environmental and social responsibility).

Customer-centric SC design principles (from Zappos):

• Remove friction from the customer journey — free shipping, free returns, 24/7 phone support
• Invest in the "trust loop" — policies that make customers comfortable buying = higher LTV
• Define yourself by service, not product — the supply chain is the brand
• Empower frontline staff to resolve issues without escalation — decentralised decision rights

Sustainability dimensions in reverse logistics:

Companies outsource to increase Supply Chain Surplus by leveraging a third party's scale, specialisation, or lower cost base. The decision should always be evaluated against the full strategic picture — not just unit cost.

Why companies outsource:

• Access to specialised capabilities or assets not worth building in-house
• Convert fixed costs to variable costs (flex capacity with demand)
• Focus internal resources on core competencies
• Exploit third-party economies of scale

TCO is the most important concept in sourcing decisions. It forces managers to look beyond the quoted unit price to the total cost of bringing a product into the supply chain and through its lifecycle.

Tailored sourcing means matching the sourcing location to the demand characteristics of each product — not applying a single global sourcing strategy to everything.

Tailored sourcing is the strategic bridge between Session 5 (push/pull, newsvendor) and Session 7 (global procurement). The critical ratio tells you how much to order; tailored sourcing tells you from where to order it.

Standard wholesale contracts create misaligned incentives: the retailer bears all demand risk, so orders conservatively; the manufacturer loses revenue from understocking. Risk-sharing contracts redistribute risk to unlock higher order quantities and joint profit.

The LEGO case is the definitive study in strategic misalignment in outsourcing. Between 1993 and 2004, LEGO's component complexity exploded — from an unknown number of parts to 3,560 unique shapes, 157 colours, and 10,900 total elements. Fill rates to retailers collapsed from disciplined levels to 5–70% variability. The company had an "operations nightmare."

Why Flextronics failed LEGO:

Hau L. Lee's 2004 HBR article argued that the prevailing obsession with supply chain speed and cost was not only insufficient — it was actively creating fragile supply chains that failed precisely when they were needed most.

The reason: efficiency-optimised supply chains are designed for steady-state conditions. They are optimised for the average demand scenario, not for volatility, disruptions, or structural market shifts. The moment reality deviates from the plan — a demand spike, a supplier failure, a geopolitical shock — a lean chain has no slack to absorb it.

Importantly, the three As are not substitutes — a supply chain needs all three. Agility without Alignment means partners react at cross-purposes. Adaptability without Agility means the firm evolves structurally but cannot respond to short-term shocks. Alignment without Adaptability means partners collaborate within a structure that becomes obsolete.

Agility is the ability to respond quickly to short-term, unexpected changes in supply or demand — and to recover from disruptions before competitors can capitalise.

Adaptability is the strategic capacity to evolve the supply chain's fundamental design as markets, technologies, and geopolitics shift permanently. Unlike agility (reacting to noise), adaptability addresses long-term signals.

Structural failures from inertia: Lucent kept its manufacturing hub in Oklahoma City while the market moved to Asia. Components shipped from Asia to the US for assembly, then shipped back to Asian customers — "frequent-flyer" components adding cost and time. More adaptable rivals used contract manufacturers in Asia to undercut Lucent's prices.

Adaptability practices:

• Monitor global trends continuously — track economic development, political shifts, demographic change, technology cycles
• Design for supply flexibility: Commonality (shared components across product lines), Postponement (standardise early stages), Standardisation (allows switching between supply networks)
• Strategic segmentation: Run different supply chains for different product lines as markets diverge
• Use intermediaries to access reliable vendors in unfamiliar markets without building a permanent presence

Alignment is the synchronisation of incentives across all supply chain partners so that when each player maximises its own interests, it also optimises the chain's overall performance. Misaligned incentives are the root cause of most coordination failures — including the Bullwhip Effect.

Alignment practices:

• Equal access to information: Provide all partners with the same forecasts, sales data, and plans — no information asymmetry
• Clarify roles and responsibilities: Avoid conflict from ambiguous boundaries
• Redefine partnership terms: Share risks, costs, and rewards equitably — this is where risk-sharing contracts (Session 7) connect
• Align incentives explicitly: Metrics and rewards must reflect chain performance, not just individual firm performance

Yossi Sheffi and James Rice (MIT) define supply chain resilience as the ability to absorb disruptions and recover to normal performance — or improve beyond it. The question is not whether disruptions will occur, but how fast the chain returns to full capacity.

Two resilience strategies:

Session 9 ties together the entire course. Here is how each session's key concept connects to the Triple-A framework and the central goal of maximising Supply Chain Surplus: