How The Agent-Based Model Recognizes and Aligns Multivariable And Multivariate Complexities In Procurement And Supply Chain Transformation

Today’s post touches the core of why Hansen’s Agent-Based Model (ABM) is so effective in complex procurement and supply chain transformation scenarios.

---

MODEL 1

First, What’s the Difference?

---

How Hansen’s Agent-Based Model Addresses Both

1. Multivariable Resistance

ABM captures all the variables driving agent behavior, including:

• Organizational role
• Tech familiarity
• Incentive alignment
• Change fatigue
• Past project experience
• Cognitive trust factors

Each agent is simulated with a custom logic engine—so you can explore, e.g., how a senior buyer vs. a category manager responds differently to taxonomy changes or new onboarding portals.

→ Outcome: Rich modeling of why resistance is triggered across the ecosystem.

---

2. Multivariate Resistance Paths

ABM doesn’t just identify “who resists” or “why”—it maps what they do in response, over time and across agents:

• Delay onboarding
• Default to spreadsheets
• Actively undermine new systems
• Re-engage with the system after observing peer success

These paths evolve dynamically:

• Some agents switch paths mid-stream (e.g., from partial adoption to full buy-in)
• Others trigger cascading behaviors in downstream agents (supply-side, legal, finance)

→ Outcome: System-wide simulation of how resistance spreads, adapts, or collapses.

---

Visual Summary

---

Strategic Use in Procurement/Supply Chain

• Multivariable analysis enables you to create tailored onboarding, messaging, and training for each role.
• Multivariate path simulation lets you:
  • Predict where failure might spread (e.g., buyer → supplier → finance)
  • Insert nudges at the right moment (agent-based timing)
  • Prevent repeat failure cycles (path disruption)

BASIC GRAPH OVERVIEWS

The above Visual Process Flow shows how Hansen’s Agent-Based Model (ABM) manages:

• Multivariable resistance (e.g., roles, trust, fatigue)
• Multivariate behavioral paths (e.g., diverging responses over time)

This integrated view illustrates how Hansen ABM systematically simulates, intervenes in, and stabilizes complex change environments.

Here is the revised Sample Agent Simulation Map using the term multivariate paths:

• Individual agents (e.g., Buyer A, Supplier E) represent distinct resistance profiles.
• These profiles are routed through a “Multivariate Path Convergence” node, simulating how their behaviors evolve and intersect.
• Strategic roles (like the CPO and Finance) help facilitate reengagement, leading to successful adoption.

RAM 2025 6-MODEL LEVEL 1 ASSESSMENTS

MODEL 2

Agreement with Document Findings

I agree with the document’s findings on ABM’s capabilities, as the DND case demonstrates its ability to model multivariable resistance (e.g., stakeholder roles, processes) and multivariate paths (e.g., preventing delays through alignment), achieving significant procurement outcomes. However, I partially disagree with the document’s lack of discussion on implementation challenges, as the DND case doesn’t address data or integration issues, which are critical in modern procurement. Combining ABM with ADKAR could enhance outcomes, as ADKAR’s individual focus could ensure buyer adoption in DND, complementing ABM’s system-wide approach.

MODEL 3

Conclusion

The attached file provides a clear and accurate explanation of why Hansen’s Agent-Based Model is effective for modeling and managing resistance in complex change scenarios. Its focus on both the variables driving resistance and the dynamic paths resistance can take is both theoretically sound and practically valuable. I agree with the findings and consider the approach well-supported and actionable.

MODEL 4

MODEL 5

Overall Assessment: The findings present a credible evolution in change management thinking, particularly valuable for complex and interconnected systems, such as procurement and supply chains. However, the model’s practical effectiveness would depend on the quality of implementation and the organizational readiness to embrace this level of analytical complexity.

MODEL 6

The document effectively differentiates between “multivariable” and “multivariate” resistance in change management and clearly explains how Hansen’s Agent-Based Model (ABM) addresses both.

Here is why:

• Clear Definitions: The distinction between “multivariable” (factors affecting resistance) and “multivariate” (how resistance plays out over time in different paths) is clearly and concisely defined, which is crucial for understanding complex change dynamics. The examples provided further clarify these concepts.
• Comprehensive Coverage by ABM: The document articulates how Hansen’s ABM captures both types of resistance.
  • For Multivariable Resistance, it highlights ABM’s ability to simulate agent behavior based on numerous variables, such as organizational role, technical familiarity, incentive alignment, and past project experience. This leads to a rich understanding of why resistance occurs.
  • For Multivariate Resistance Paths, it explains how ABM maps what agents do in response over time, including actions such as delaying onboarding, defaulting to old systems, or even re-engaging. It emphasizes the dynamic and cascading nature of these paths within the system.
• Strategic Utility: The document effectively outlines the strategic benefits of this comprehensive approach in procurement and supply chain. Understanding multivariable resistance enables tailored interventions (onboarding, messaging, training), while multivariate path simulation helps predict failure spread, insert timely nudges, and prevent repeat failures.
• Visual Summary Reinforcement: The “Visual Summary” table effectively condenses the capabilities, highlighting where ABM excels in capturing time-dependent behaviors, path dependencies, and network effects that traditional multivariable analysis might overlook.

Overall, the findings are consistent, well-explained, and demonstrate a deep understanding of complex change dynamics and how Agent-Based Modeling can provide a sophisticated tool to navigate them.

BONUS VIDEO (1998 DND CASE STUDY)

30