In an era where a single shock can ripple through global supply chains, Tejaskumar Vaidya has earned a reputation for turning operational chaos into calm, repeatable performance. His signature Planning systems that respect real-world constraints and still move fast. “Optimization is easy,” he tells me. “Executable plans, plans that human teams trust, are the hard part.”
This feature presents Vaidya’s work and thinking in two halves: first, the playbook he applies across complex manufacturing networks; second, highlights from two recent research papers that translate those ideas into structured, peer-reviewed guidance.
Why It Matters
Leaders don’t hire Vaidya to install a tool. They bring him in when “paper-perfect plans” crumble on the shop floor, when schedules buckle under finite capacity, when inventory swells to buy stability, when every change request triggers a week of rescheduling. His response is refreshingly direct: model reality, expose trade-offs, and close the loop between what was planned and what actually happened.
“Most failures aren’t algorithmic,” he says. “They’re governance failures such as unclear policies, business practices, hidden constraints, or no mechanism for learning from overrides. Fix those, and even a modest planning engine can look brilliant.”
Inside the Playbook
1) Feasible-First Modeling
Vaidya begins by aligning data, policies, and constraints so the system stops “promising the impossible.” That means modeling true line rates and changeover times, honoring lead-time variability, and locking in realistic buffers. The test he uses is simple: “If a plan can’t be executed without heroics, it isn’t a plan.”
2) Scenario Transparency
Instead of arguing from instinct, leadership sees structured trade-offs: service vs. cost vs. inventory. Scenario runs surface the consequences of each choice, accelerating decisions and reducing last-minute gyrations. “Great planners aren’t fortune tellers,” he says. “They’re curators of choices.”
3) Human-in-the-Loop Workflows
Planners can override the system and the reasons for those overrides are captured as structured data. Those signals flow back into models and policies, shrinking the gap between theory and practice. “If the plan and the floor disagree, we don’t punish the floor,” Vaidya adds. “We learn why.”
4) Variance Control
He insists on eliminating silent process deviations: ad-hoc lead-time padding, rogue routings, and shadow spreadsheets. The outcome is a simpler, more predictable rhythm. “Stability is a performance multiplier,” he notes. “Once you have it, everything else compounds.”
5) Explainability as a Feature
Whether it’s a scheduling move or a stocking decision, stakeholders get the “why.” That clarity reduces friction, speeds approvals, and builds long-term trust in the system.
Two Papers, One Philosophy
Vaidya’s research expands these ideas in practical, practitioner-ready ways, no vendor specifics required. The first paper explores digital-twin-driven production planning; the second lays out an integrated resilience framework for planning and execution.
Paper One: Digital Twin–Driven Production Planning
The Problem
Traditional planning is episodic and deterministic; factories are continuous and stochastic. Between planning cycles, conditions drift (machines slow, quality varies, rush orders appear) and static schedules underperform.
The Proposal
A digital twin mirrors the planning and production systems in near-real time. It ingests telemetry and shop-floor signals, forecasts the near future (throughput dips, potential bottlenecks, quality risks), and feeds those insights back into the planning layer. The result is an adaptive loop: plan – sense – predict – adjust, with humans in the loop to validate and prioritize changes.
What Changes on the Ground
- Schedules are adjusted proactively rather than reactively.
- Maintenance and quality signals become planning signals.
- Planning cycles compress, not by rushing people, but by reducing uncertainty before it hits the line.
Vaidya is careful to emphasize governance over gadgets. “A twin isn’t a toy model,” he says. “It’s an accountability model. If the twin says the line will slow at 3 p.m., the system must show who sees that, what action they can take, and how fast the plan adapts.”
Takeaway for Leaders
Start with a narrow slice, one family of products, one critical line, then scale. The win isn’t merely a better forecast; it’s a shorter sense-to-decision loop that cushions volatility without resorting to brute-force inventory.
Paper Two: Enhancing Supply Chain Resilience with Integrated Planning
The Problem
Many organizations still live with a split brain: tactical planning over here, execution over there. Disturbances such as supplier hiccups, transport delays, demand spikes arrive faster than siloed processes can respond.
The Proposal
Vaidya’s second paper argues for a synchronized framework where planning and execution share a continuous feedback lane. Exceptions propagate automatically, model horizons and policies adjust by rule, and replanning happens by design rather than by emergency.
What Changes on the Ground
- Disruption scenarios are pre-built and measurable (response time, visibility, adaptability, recovery).
- Replanning triggers are explicit, no detective work required.
- Decision rights are clarified: who approves a policy shift, who executes a schedule change, who validates recovery.
“Resilience isn’t a dashboard,” Vaidya reminds me. “It’s a choreography. You get better not because you have more screens, but because your teams know exactly what to do when the music changes.”
Takeaway for Leaders
Before chasing new technologies, harmonize the ones you have. A well-governed connection between planning and execution yields outsized resilience, often with the systems already in place.
What’s Next
He’s focused on making planning “adaptive by design”, pulling in upstream risk signals, sustainability limits, and demand micro-shifts without destabilizing operations. The guiding principle won’t change: feasibility first, transparency always, learning as standard work.
“Volatility isn’t going away,” Vaidya concludes. “But panic can. If we close the loop between sensing and deciding, and keep humans in that loop, operations become not just efficient, but resilient.”
_________
