Sales and operations planning works only when its subprocesses are integrated and complemented by temporary planning during crises; AI promises automation and better scenarios in S&OP but delivers value only with complementary organizational and technology changes.

Main Finding

Sahil Ahmed’s thesis (2026) shows that the tactical Sales & Operations Planning (S&OP) process is a dynamic, multi‑subprocess system that must be designed contingently: integration needs differ across S&OP subprocesses, organizational resilience during disruptions emerges from interplay between routine S&OP and temporary planning processes, and artificial intelligence (AI) is most effective when it augments—not simply replaces—human and organizational capabilities. The thesis develops conceptual frameworks (subprocess integration, transition capabilities for supply‑chain planning, and a CIMO‑based model for AI‑enabled S&OP) and synthesizes mechanisms and contextual conditions that determine successful adaptation of S&OP to integration, resilience, and AI challenges.

Key Points

• Integration is multi‑dimensional and situational:
  • Integration in S&OP is achieved through coordination, collaboration, and alignment, but the required mix varies by subprocess and planning situation. A one‑size‑fits‑all S&OP design is insufficient.
  • The thesis proposes a subprocess‑level integration framework and an integration maturity model (showing different requirements during routine vs. disrupted states).
• Resilience arises from process configuration and temporary planning:
  • Organizational resilience during supply‑chain disruptions emerges from the interaction of routine SCP (including S&OP and S&OE) and temporary SCP processes that are activated in disruptive states.
  • The research identifies practices and transition capabilities that allow organizations to shift from routine to disruption‑mode planning (e.g., faster decision loops, ad‑hoc cross‑functional teams, reallocation of decision rights).
• AI enables but requires careful implementation:
  • AI can augment many S&OP activities (forecasting, scenario simulation, decision support, automated recommendations), but benefits depend on data quality, model transparency, human trust, governance, and organizational capabilities.
  • The thesis conceptualizes “AI‑enabled S&OP” and uses a Context–Intervention–Mechanism–Outcome (CIMO) lens to explain when AI implementations succeed (mechanisms such as automation of routine tasks, improved decision speed/quality, and augmented human judgment) and what organizational conditions (HOT: Human, Organization, Technology) are required.
  • Common barriers: poor data, lack of integration across systems, low user trust, skills gaps, and misaligned incentives.
• Practical synthesis:
  • The studies combine to produce actionable guidance for designing, adapting, and improving S&OP processes so they remain effective under varying contexts (normal operations, disruption, and digital transformation).

Data & Methods

• Overall approach: Qualitative multiple case‑study research across three empirical studies that together underpin five appended papers (I–V). The thesis uses comparative case analysis to develop theory and practical guidance.
• Empirical material: In‑depth case studies of multiple industrial firms (several organizations across sectors), drawing on interviews, internal documents, observations, and archival data. (Specific firm identities and detailed case counts are reported in the thesis’ methodology section and appended papers.)
• Analytical techniques:
  • Thematic and cross‑case coding to identify patterns across subprocesses, resilience practices, and AI implementations.
  • Process tracing of transitions between routine and disruptive planning modes.
  • CIMO‑based synthesis to map contextual conditions to interventions (AI and process changes), the mechanisms they trigger, and observed outcomes.
• Outputs: Conceptual frameworks and maturity models (integration maturity; transition capabilities), typologies of temporary SCP processes, and a CIMO‑driven research agenda for AI in S&OP. Several papers were presented at relevant conferences and are under journal review or in preparation.

Implications for AI Economics

• Productivity and task‑level complementarity
  • AI in S&OP is likely to raise productivity mainly through task‑level augmentation (forecasting accuracy, scenario evaluation, faster decision cycles) rather than wholesale replacement of planners. Economists should model complementarity between AI tools and human judgment when estimating productivity gains.
• Heterogeneous firm gains and inequality
  • Benefits are contingent on firms’ data quality, integration maturity, and human capital (HOT capabilities). This implies heterogeneous returns to AI adoption across firms—firms with superior data and organizational readiness will capture disproportionate gains, potentially widening productivity and market share gaps.
• Labor effects and skill upgrading
  • Planners’ roles will shift toward overseeing models, handling exceptions, strategic decision making, and cross‑functional coordination. Labor demand is likely to shift from routine forecasting tasks toward roles requiring judgment, interpretation, and governance—implying upskilling rather than pure job loss in many contexts.
• Adoption barriers and economic frictions
  • Non‑technical frictions (trust, governance, incentives, coordination costs) can delay or blunt the economic returns from AI. Economic analyses should account for adoption lags and organizational adjustment costs when estimating aggregate impacts.
• Impact on resilience and systemic risk
  • AI may both strengthen and weaken resilience: it can improve early detection, scenario planning, and responsiveness, but over‑reliance on models, shared data biases, or simultaneous automated responses across firms could produce systemic vulnerabilities. Macroeconomic and network models should incorporate these second‑order effects.
• Market structure and investment dynamics
  • Firms may invest more in data infrastructure, integration platforms, and AI governance—raising fixed costs and potentially increasing scale advantages. This dynamic is relevant for studies of concentration, market entry, and competitive advantage.
• Research opportunities and empirical strategies for economists
  • Causal evaluation: Use staggered rollouts, difference‑in‑differences, synthetic controls, or instrumental variables around AI tool deployments in S&OP to estimate productivity and resilience effects.
  • Microdata: Combine firm‑level S&OP adoption indicators with transaction, inventory, and performance metrics to estimate effects on service levels, working capital, stockouts, and lead times.
  • Structural/behavioral models: Quantify how AI changes planners’ decision policies and the value of information-sharing in contracting settings.
  • Network/aggregate effects: Study spillovers and systemic risk by modeling correlated AI adoption across suppliers/customers and its effect on supply‑chain fragility.
• Policy and governance considerations
  • Regulators and firm managers should consider standards for explainability, model validation, and data sharing to ensure robust, equitable benefits from AI in supply chains. Economic policy questions include supporting upskilling, managing concentration risks, and addressing systemic resilience.

Suggested immediate empirical questions for AI economists: - What is the causal impact of AI adoption in S&OP on firm‑level inventory days, service levels, and sales volatility? - How do returns to AI investment vary with pre‑existing IT integration and organizational maturity? - Does AI deployment in planning reduce the cost of responding to disruptions, and under what conditions does it increase systemic fragility?

Summary This thesis provides a detailed, empirically grounded perspective showing S&OP must be designed contingently across subprocesses, that resilience emerges from coordinated routine and temporary planning, and that AI is a powerful but context‑dependent augmenting technology. For AI economists, the work highlights rich empirical variation and a set of concrete mechanisms and contextual moderators (data, governance, skills) to test when estimating AI’s micro and macroeconomic effects in supply chains.