The optimization center of excellence @ BLEND 360 leads our organization in developing perspective analytics, mathematical programming, and optimization solutions for our clients in a wide variety of industries. We are also committed to share our experience and innovative solutions with the larger optimization community because we believe knowledge sharing is an integral part of innovation.

Recent Posts

Infeasibility resolution with PuLP

12 minute read

Resolving infeasibility of large problems depends on identifying the conflicting constraints which is not a simple task. A problem could have multiple sources of infeasibility that may or may not overlap. Relaxing a problem’s constraints always resolves the infeasibility but it would not identify the source necessarily since it always depends on how much penalty is assigned to each relaxation coefficient. The usual optimization solvers also do not have a perfect way of resolving infeasibility since the exact method of identifying the source of infeasibility and then determining the minimum number of constraints that have to be removed is a problem that can end up being more complex than the original problem.

Dynamic transshipment policies for interstate moving companies with load uncertainty

26 minute read

Interstate moving companies move household items from one location to another. They often optimize their routes by assigning a truck to n > 1 households in a single day within a region, move the items to a storage location and then solve a dispatch problem to deliver each household’s items from storage to the new location within a time window. A crucial step in planning is knowing in advance the size of the cargo that will have to be picked up. This was traditionally done through an examination by an agent before the scheduled move. With recent changes due to COVID restrictions, moving companies rely on customer’s estimate of volume and size of their items. These estimates often have errors and may cause order cancellations at the last minute, either due to price difference from the initial quote or limited capacity of the assigned truck.

Optimal staff level using dynamic job shop scheduling

14 minute read

A company operates bases in the US where vehicles depart from and arrive to. Very broadly, this company provides several types of services. For their operation, they need to staff the bases with technicians and customer representatives. The primary goal is to improve how they staff by aligning it better with the volume of arriving and departing vehicles. With this, they are hoping to reduce under- and over-staffed moments, improve customer service, and reduce labor costs.