The "Pipeline" Project is a study about issues associated with the supply and demand for mathematics students and personnel in educational institutions and the workplace.  The study is coordinated by ICMI at the request of its mother organization, the International Mathematical Union (IMU). It also builds on, and is an extension of, the work of the Survey Team for ICME-11 on the topic of "Recruitment, entrance and retention of students to university mathematical studies in different countries" and chaired by Derek Holton. The study aims at providing data for decision making in different countries as well as promoting better understanding of the situation internationally.  At the moment, the following countries are participating in the study: Australia, England/UK, Finland, France, Korea, New Zealand, Portugal, and USA.

Responses to a preliminary questionnaire from a group of countries showed that the situations in different countries differed markedly, and it might not be possible to identify "global trends" as originally envisaged. Given the difficulty of collecting comparable data, it was agreed that instead of a traditional comparative study, the project should be conceived of as a series of national case studies centered around four crucial transition points:

1. From school to undergraduate program
2. From undergraduate program to teacher education (and to school teaching)
3. From undergraduate program to higher degrees in mathematics
4. From higher degrees to the workforce. Workforce under study includes school teachers, research mathematicians, mathematically intense professionals, and workers in other careers suitably defined for each country.

Participating countries have been invited to collect data in answering the questions below. 

• How many applicants are there at each transition point?
• What are the transition or admission requirements?
• How many applicants meet the admission requirements?
• How many applicants are successful in the transition (i.e., number of applicants admitted)?
• What are the actual credentials or qualifications of those admitted to the course?
• How many students complete that stage?
• For students who complete that stage, what percentages of them go into different streams in higher educational institutions or the workplace?

As far as possible, time-series data are collected so that trends can be identified. The research questions for different countries may differ slightly, and since the data available across countries may not be equivalent, the kind of data used to answer the research questions may also be different according to the situations in the countries concerned. The data-set utilized in each country will be specified, and the terms used precisely defined. As far as possible, common definition of terms would be established, but where terms mean different things in different countries, they will be clarified on a national basis so that the results of the study are understood by readers outside the country.

Members of the project have met and shared their initial findings at ICME-11, and further data collection and analysis took place thereafter till the end of 2008. A report on the project was presented at ICM-2010 in Hyderabad, India.