Background of Pharmaceutical Scheduling Challenge
This describes a scheduling implementation for a pharmaceutical manufacturer of vaccines. The complexity of the schedule stems from the length of the scheduling horizon, number of operations, variable times taken by the operations, and availability of both primary and auxiliary resources. Eyelit Technologies is used to generate / regenerate buildable schedules for vaccine production.
Pharmaceutical Scheduling Requirements
Vaccine production requires the coordinated scheduling of many operations. In turn, each operation uses a particular workstation or production area along with auxiliary resources such as operators (skill level) and other workstations. The schedule needs to meet all the following requirements:
- Ensure all delivery / due date commitments are met
- Minimize changeover times between batches
- Limit the number of batch starts per week to match plant capacity
- An operation cannot be initiated unless all required resources are available, but resources can be held and released at different times
- Ensure that staffing levels are maintained, while conforming to an employee “skill matrix”
- Schedule preventive maintenance in the most suitable time slot
Using the Eyelit Technologies routing definition interface, routings are set up which capture details such as operation sequence, workstation by operation, auxiliary resources by operation, and operation cycle time. Operators are grouped into 2 sets: Operators1_9, and Operators 10_37, depending on skill sets. The Operator set is associated with an operation, based on the skills required.
The number of operators required for an operation from the Operator set are also defined using the Resource requirements for each operation.
Changeover times between batches are modeled through the Changeover Matrix.
Results of Implementing Pharmaceutical Multi-Operation Scheduling Solution
The Eyelit Technologies multi-operation scheduling functionality considers all primary resources and auxiliary resources together to build a schedule that is optimal for the entire facility. A legacy or traditional approach would be to create workstation wise schedules which result in optimality at each workstation but that results in overall sub-optimal solutions for the facility.
The schedule considers every primary and auxiliary resource. Batches are scheduled based on availability, with the aim of minimizing the make-span. Every job for each batch on each machine can be tracked, and the entire process for a batch can be visualized. The schedule considers operator availability as required for the operation. SM_01 through SM_37 represent operators, and their assignment to an operation by batch number.
Eyelit Technologies can holistically optimize across multiple operations and schedule all batches so that all delivery due date requirements are satisfied. Further the schedule minimizes changeover times and reduces overall makespan, while ensuring that staffing assignments adhere to the skill matrix. In this example, minimization of changeovers resulted in a makespan reduction of 19%.
A line block is added at the optimal time, to support preventive maintenance. The schedules are generated within a few minutes, allowing the rapid regeneration of schedules as required.