Summary of Generative Discrete Event Process Simulation For Hidden Markov Models to Predict Competitor Time-to-market, by Nandakishore Santhi et al.
Generative Discrete Event Process Simulation for Hidden Markov Models to Predict Competitor Time-to-Market
by Nandakishore Santhi, Stephan Eidenbenz, Brian Key, George Tompkins
First submitted to arxiv on: 6 Nov 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The proposed research studies the challenge of predicting when a competitor product will be available to customers, with the goal of revising this estimate as new information becomes available. The scenario involves Firm A, which is competing against Firm B in the same industry, gaining periodic insights into Firm B’s activities by observing their resource usage. To achieve this, Firm A builds a model that leverages knowledge of the underlying processes and required resources to predict when Firm B will be ready to sell its product. This model uses a Parallel Discrete Simulation (PDES) process model as a generative model to train a Hidden Markov Model (HMM). The study examines how many resource observations are needed to accurately assess the current state of development at Firm B, considering factors such as process graph densities and resource-activity map densities. The results show that the HMM achieves a prediction accuracy of 70-80% after 20 daily observations of a production process that lasts 150 days on average. This research provides insights into the level of market knowledge required for accurate early time-to-market prediction. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper explores how to predict when a new product will be available, using information about its development process. Imagine you’re trying to guess when your competitor will release their new car model or high-capacity battery. The researchers propose a way to make this prediction by analyzing the resources they use during development. They show that it’s possible to build a model that can predict the time-to-market with an accuracy of 70-80% after observing the development process for about 20 days. This research helps us understand how much information is needed to make accurate predictions. |
Keywords
» Artificial intelligence » Generative model » Hidden markov model
