Machine Learning for Operations Inc. (MLO) is a consulting firm that develops customized machine learning software to support operational decisions (e.g., for forecasting, procurement, scheduling, etc.). On average, MLO receives 1.6 new project requests per week. These arrivals follow a Poisson process.

On each project, the firms consultants work jointly in teams of two a domain expert and a modeler. The same two individuals work as a team throughout the year. On average, each project requires one team working, on average, for 50 business days (from the time the consultants begin working on the project to handing over the results to the client). Historical data reveals that the standard deviation of project work time is 45 business days. MLO assigns projects in first-come-first-served order to the first available consultant team. Each team works on at most one project at a time.

MLO currently employs 36 consultants, working as 18 teams. The firm pays each consultant $ 12,000 per month in salary and benefits, whether or not they are working on a project. On average, MLO charges $ 150,000 per project, based on its consulting rate of $ 3,000 per day of work on the project. Assume that each week consists of five business days, every month has four working weeks, and consultants work for all 12 months in a year.

OpsAI Inc. is another firm that does work very similar to that of MLO, but in a different geographical region (so they do not compete directly). Both firms are organized along similar lines with teams of two consultants working on each project, comparable project work times (same mean and standard deviation of work time per project), and the same approach for quoting promise time. OpsAI is a somewhat smaller operation than MLO; on average, it receives a project request from a client once every 6 business days, with a coefficient of variation of 1. It charges clients the same rate as MLO.

Consulting firms closely monitor their cost-efficiency which is the total monthly salaries for consultants expressed as a % of the firms total monthly revenue. What is the cost-efficiency of MLO (with its proposed 90% billable target)? Of Ops AI?

Compare the customer responsiveness and cost-efficiency of the two firms, and comment on the factors that underlie the differences. Are there strategies that either firm can adopt to maintain good cost-efficiency (low values preferred) while also ensuring good customer responsiveness? Explain.