A game is a decision that someone is trying to answer using GridRank. Setting up a game is the responsibility of the Game Master. Creating a game involves entering in some basic information, such as the name of the game, a label name for criteria, a label name for options, a label name for the aggregate criteria, and the game type (Pick One, Pick Many, or Weighted Criteria). The game types are further described below:
- Pick One Game
The Pick One game is a type of game where the goal is to select the most desired option from the set of options. A real-world example of this type of game might be selecting the best place for a summer vacation. In GridRank, the system calculates the single option that best meets the evaluation criteria for that player. GridRank also calculates the ratio of value to cost and identifies the best value choice by combining as the highest ratio of value to cost.
- Pick One Game
The Pick Many game is a type of game where the goal is to select as many options as possible within a given constraint, such as a budget. A real-world example of this type of game might be selecting a group of projects that can be implemented in this year’s budget. In this game, the Game Master establishes a cost constraint for the game and a cost for each option and a goal for the game – either maximize or minimize. GridRank uses a linear programming algorithm to optimally determine the set of options that best meet the evaluation criteria subject to a constraint for that player.
- Weighted Criteria Game
The Weighted Criteria game is a type of game where the player is trying to identify the weighting of their preferences. A real-world example of this type of game might be identifying what is the most important factor or evaluation criteria when selecting a university, such as academics reputation, social life, athletic programs, and diversity. The resulting weights from a weighted criteria game are often used as inputs into a Pick One or Pick Many game; in what we sometimes refer to as a two-pass game.
To read more about terminology used in GridRank, click here