무제 문서

Fly RISKOptimizer to Profit Maximisation

Here's a typical optimisation problem that benefits from RISKOptimizer's powerful combination of optimisation and simulation:

Piedmont Commuter wants to use yield management to identify the optimal number of full and discount fare seats to sell on a given flight. It also needs to identify the optimal number of reservations to accept in excess of the number of available seats - the classic "overbooking" problem.

There's just one catch to this standard optimisation problem -- some estimates in the model are stochastic. This includes the number of passengers that will actually show up to board the flight, the number of reservations that will be demanded in each fare category and the cost of bumping a passenger (i.e., sometimes a $150 travel voucher will suffice, while sometimes a free round trip is necessary).

Piedmont yield management model using
RISKOptimizer. Click to view close-up.

Traditionally, single point estimates are used for these items, allowing a normal optimisation to be performed. But what if your estimates weren't right? You might end up taking too few reservations, sending seats out empty, or overbooking too much. You also might sell too many discount seats -- lowering your profit, or setting aside too many full fare seats, resulting in half-filled planes. RISKOptimizer will solve this optimisation problem while allowing you to account for the uncertainty inherent to your model!

With RISKOptimizer, a probability distribution can be assigned to each uncertain value in the yield management model, as shown in the graphic at right. This includes 1) the % of "no-shows" for full and discount fare reservations, 2) the number of reservations demanded in each fare class and 3) the cost of bumping. The cells that RISKOptimizer will adjust ("the changing cells") are 1) the total number of reservations that will be accepted and 2) the % of the total reservations taken that will be set aside for full fare seats. Remember, you'll take more reservations than you have seats because of the "no-show" problem.

The objective of the optimisation is to maximise the mean of the distribution of the simulation results for Profit. But you have a couple of constraints. First, Profit can never drop below zero, in any circumstance. This is done by adding an "iteration" constraint that Profit>0 for all iterations of all simulations run. You also need to keep an acceptable level of risk for Profit levels. This is done by adding a "simulation" constraint that the standard deviation of the distribution for Profit must be less than 400. This constraint is checked at the end of each simulation run.

Understand the risk associated with your optimal
decision using RISKOptimizer. Click to view close-up.

As the optimisation starts, RISKOptimizer generates new combinations of values for the total number of reservations and the % of full fare reservations. For each combination it runs a simulation. During each simulation it runs thousands of iterations, sampling values from the entered probability distributions and generating a probability distribution for Profit. At the end of the optimisation the values for the total number of reservations and the % of full fare reservations that maximise the mean of the distribution for Profit, while meeting your constraints, are found.

This is a big advance over the optimisations you would have performed before RISKOptimizer. First, instead of making assumptions about your uncertain inputs, you model them using probability distributions. Secondly, your optimisation results factor in that uncertainty -- instead of assuming it away. You get detailed information on the risk associated with your results so you can make a more informed decision!

Product Versions
RISKOptimizer has a limit of 80 adjustable cells, with unlimited probability distributions. RISKOptimizer Industrial features unlimited adjustable cells and unlimited probability distributions.

주문 안내

가격에는 부가세를 제외한 1년간의 Maintenance를 비롯한 모든 제반 비용이 포함되어 있습니다. 필요하시면 반품 및 환불에 관한 규정 과 배송 방법 및 기간 를 참고 바랍니다.

RISKOptimizer Industrial

Price: $2,095.00

RISKOptimizer Standard

Price: $1,495.00

RISKOptimizer System Requirements
Minimum Platform: IBM PC compatible Pentium-equivalent or higher, 16MB RAM, Windows 98, NT 4.0, Windows 2000, Windows XP
Recommended: 32 MB RAM, or greater
Spreadsheet: Windows Excel 97 or higher
Version: 1.0
Developer's Kit : RISKOptimizer Developer's Kit
Technical Support: 30 days free. Further support available through Maintenance Plan.
Demo: Web download and free demo CD with trial version available.
Training: Available through Palisade's Software Training Courses.
Recommended Books: Decision Making Under Uncertainty with RISKOptimizer; Financial Models Using Simulation and Optimization; Financial Models Using Simulation and Optimization II; RISKOptimizer for Business Applications; Trends and Tools in Operations Management