Start a project

The Question

How do you plan port labor when demand and availability keep shifting?

Case study · Maritime · Workforce dispatch

ConfidentialClient work product

A year of dispatch, off the physical board

Client
(withheld)
Industry
Maritime: port labor dispatch
Decision type
Staffing and scheduling
Model
Dispatch simulation linked to an availability model trained on years of history
Horizon
A year of dispatch, every assignment
Finding
Hundreds of thousands of dollars saved per year

Client's brief · reconstructed excerpt · values held in confidence

(withheld) dispatches the workforce for port work in (withheld). Every job is assigned under strict rules, and each assignment changes who is left for the next one. For years that work has run by hand, on a physical board. Then the pandemic pushed demand two ways at once, more container traffic and far fewer cruise ships, while staffing ran short. The organization needs to see a year ahead: what shifts in worker availability and in demand will do to the jobs it can fill.

Bars cover values SimWell holds and is not releasing. The released figures stand as printed.

§ 1

The decision

Dispatch has a cascade effect. A worker sent to one job is not there for the next, and the rules decide who is next in line, so every assignment reshapes the ones after it. Across a full workforce and a year of ship calls, nobody can see those interactions on a board.

The client wanted to test changes in worker availability and in demand over a full year, and see the effect on service and cost.

§ 2

What the model did

SimWell built a simulation of the entire dispatch process and linked it to an availability model. Trained on several years of historical data, the availability model predicts whether each employee will be available on a given day, and the simulation checks it before every dispatch.

The output is every dispatch decision for the period: which job, which worker, when. That runs to hundreds of thousands of lines. The client analyzed it in a BI tool against its own history, with the KPIs in the same format, to confirm the model behaved like the real operation before trusting its scenarios.

Exhibit A · One day's cascade, then a year of it
Dispatch model · Schematic · Scales illustrative Dispatch lines generated 0 One day's dispatch Jobs, in dispatch order Workers, in dispatch priority Job 5 took the last available crane operator. Job 6 goes unfilled. That is the cascade. A year of dispatch, two scenarios Unfilled shifts per week Week 1Week 52 Current demand Containers up, cruises down Job Worker Predicted absent Unfilled

Cleared for release

Results

The client now tests staffing and demand scenarios months and years ahead in a few clicks. The initial result saved hundreds of thousands of dollars a year and replaced the manual work of dispatching on a physical board.

The team can run the model on its own and keep testing scenarios after the project.

SimWell helps operators make better decisions under uncertainty. We build simulation and optimization models that reflect operational reality, test scenarios before committing resources, and transfer capability so teams can run decisions independently.

Start a project

Client confidential. Client name and region withheld. Released figures stand as printed.