MobiDoc initiative in Tunisia

MobiDoc initiative in Tunisia

A mechanism for doctoral students and post-doctoral fellows to conduct research in an industrial or service company.

By Atidel B. Hadj-Alouane

Tunisia is a small North African country perhaps best known today as the birthplace of the “Arab Spring.” Tunisia, a rising democracy, is becoming a start-up hub in the North African region. The country faces various challenges, mainly economic and political issues, but what interests me as a college professor is the higher educational system that has not changed much since the Arab Spring and the associated events of 2011.

To give some background, I had the chance to see two different sides of the world when I was a completing my education. I completed my engineering degree in Tunis and received my Ph.D. from the University of Michigan in the United States. Needless to say, I have noticed many differences between both countries’ educational systems. One major difference is that Tunisian education is free. This leads to the vast majority of citizens being able to attend universities, but it also poses numerous problems to the system as a whole such as the lack of funding.

For example, consider the case of Ph.D. students in Tunisia. The lack of funding causes major hurdles for students seeking to pursue a Ph.D. because they are generally unable to acquire the appropriate amount of funding needed to do research. As a result, many students end up seeking funds on their own, which is not an easy task. In other cases, students join the industry as full-time employees, hindering their research and academic commitments. However, an on-going initiative is successfully helping these students accomplish their research goals by mitigating most financial problems.

Mobility Mechanism for Doctoral Students

Tanneries & Megisseries of Maghreb (TMM) is the largest tannery company in Africa and a supplier of high-quality finished leather to some of the biggest fashion houses in the world. Source: TMM

This initiative is called MobiDoc, and it is one of the flagship activities of the PASRI program (Project to Support the Research and Innovation System). The program is funded by the European Union and came as a response to the weak economic performance of production systems in Tunisia, which is partly due to the lack of partnership between the industry and the university, at the research and innovation levels.

The purpose of MobiDoc is to set up a mobility mechanism for doctoral students and post-doctoral fellows to conduct research in an industrial or service company. The program provides funding to candidates in the form of a stipend for a period of 36 months, with a minimum contribution of 20 percent from the company. In Tunisia, the MobiDoc framework is managed by the ANPR (National Agency for Promoting Scientific Research), which is a public organization created in 2008 and placed under the Ministry of Higher Education and Scientific Research.

Each MobiDoc instance involves a tripartite agreement binding the school (where the candidate enrolls), the ANPR and a recipient organization (the company). It describes the research project, subject of the agreement, work conditions, modalities of supervision, etc. In particular, for doctoral students, there are two advisors: a scientific advisor and a professional advisor. The scientific advisor, a faculty member of the school, plays the same customary role of student guidance to ensure the scientific quality of the work and to encourage dissemination of important findings mainly through publications. The professional advisor, on the other hand, monitors the work quality in terms of its immediate socio-economic impact on the company, such as increased productivity, sales, market shares or in some situations, better societal or environmental performances.  

What makes the MobiDoc initiative attractive is the win-win relationship established between the student and the company. For the student, it provides a comfortable financial situation and favorable conditions for success. Through the direct involvement of the company, the student gets the chance to grasp real-world constraints and issues and to better understand the difficulties of implementing realistic solutions to problems in the industry. This also strengthens the student’s professional skills, and at the same time validates his or her work academically within the scientific community. For the company, it provides further development and innovation, which is often lacking in countries like Tunisia.

There is a third winner in this experience: the research laboratory and the university as a whole. Through MobiDoc, research labs are now able to attract outstanding candidates, particularly engineers, who otherwise will turn to full-time jobs in private companies to thrive financially, even if they are deeply inclined to pursue a research career. Research labs have better opportunities to develop what is called research-actions projects that better respond to real needs and at the same time yield high-quality publications.

MobiDoc Experiences at TMM

To illustrate the MobiDoc initiative, I would like to share one of the ongoing MobiDoc experiences that is conducted by my student Mohamed Frihat, at an industrial company from the leather tanning sector: Tanneries & Megisseries of Maghreb (TMM), the biggest tannery company in Africa and a supplier of high-quality finished leather to some of the biggest fashion houses in the world. Mohamed is an industrial engineer who was working full time in TMM as an assistant production manager before he decided to enroll as Ph.D. candidate at the National Engineering School of Tunis (ENIT) and to become a member of the laboratory for Optimization and Analysis of Service and Industrial Systems (OASIS). His thesis deals with resource allocation and production scheduling in the complex manufacturing environment of TMM that is characterized by various aspects such as:

  • human resources constraints: employees’ availabilities, skills and labor legislations;
  • technological constraints: time lags between operations and set-up times;
  • random events: machine failures, absenteeism and quality nonconformance; and
  • multiple criteria: makespan, tardiness and labor cost.

At the beginning, the student considered the integrated deterministic problem of employee timetabling and hybrid job-shop scheduling. After presenting and formulating the problem as a mixed integer program, with the objective being the total labor cost, a resolution method based on decomposition and cut generation was developed. In this approach, the master problem is the employee timetabling, and the slave problem consists of task scheduling. These sub-problems were solved using ILOG MIP and constraint programming, respectively.

Given the complexity of the integrated problem, the implementation of this approach and its applicability turned out to be very difficult. In addition, the assumptions that were made, particularly the deterministic context, resulted in planning and decisions that were quite inapplicable. However, working with the above models allowed us to better understand the problem and to pinpoint key issues such as the emphasis on flexibility and robustness when developing the decision-making tool.

In the course of this work, there was a slight shift in the objectives. At the beginning of the project, the company placed more importance on labor cost. However, as the demand increased, it became more important to meet customer deadlines and to optimize the scheduling makespan. In order to adjust to this new context, in addition to capturing various uncertainty sources and perturbations, new research directions were sought which led to the following actions:  

  • inverting the sub-problems so that scheduling becomes master with more than one criteria and timetabling becomes the slave;
  • modeling various types of uncertainties and perturbation using fuzzy logic;
  • proposing different temporal robustness measures;
  • developing robust constructive heuristics and metaheuristics integrating more than one criteria; and
  • proposing a reactive scheduling approach that integrates operators’ availabilities and skills.

In addition to the above, the student developed a flexible scheduling approach where groups of interchangeable tasks are sequenced instead of single ones. This is particularly suitable for the company’s manufacturing system, where the execution order of certain tasks is hard to fix in advance. Overall, the student noticed that reactive approaches are easier to implement and present more flexibility in running the production system whether in dealing with uncertainties due to human errors, quality nonconformance or with the variable processing times due to the variability of the quality of raw material.  

Advantages and Limitations

Based on the above thesis and other experiences in service companies (one in a private clinic and another in a consulting firm), I believe that MobiDoc offers two advantages, in addition to the financial support for the student:

Resource availability: Being a partner, the company is compelled to provide all necessary means to ensure steady progress and the success of the research work. For example, for Mohamed’s thesis, TMM has invested in the development of a real-time production tracking system in order to collect the most accurate data and monitor the system performance while implementing the solutions proposed by Mohamed. For this purpose, one engineer was hired and five projects with specific tasks were proposed to senior engineering students at ENIT.

Teamwork: This is a very fundamental aspect when working in a company. In order to guarantee fast and adequate results, the student has to exchange ideas and discuss several proposals with decision-makers and employees at various levels, such as the floor manager, technicians, operators, interns, etc. This has the advantage of viewing and evaluating issues from different angles and thus correctly setting the priorities.

In contrast, some constraints and limitations were raised by some students who are conducting MobiDoc theses, such as:

The need for short-term solutions: Although there is a certain understanding of the nature of research projects, companies tend to expect tangible results very early on during the span of the thesis period. This will sometimes push the student to juggle between methodological approaches and easily implementable solutions until he or she finds the right compromise. Sometimes, this process hinders the academic work progress, especially when students are required by their school to publish their work in conferences and journals, which in turn requires original and high-quality research. It seems that there is a challenge in reaching a common ground where high-quality research fits in both the company’s and the university’s standards.

Occasional change of research objectives: As the company adjusts to its socio-economic environment, a MobiDoc project may have to adjust its objectives as well. Indeed, any change in the company policy such as an increase or a decrease of its capital or market share may influence the research orientation of a thesis.

In conclusion, it is important to note that a MobiDoc thesis is subject to the same academic requirements as any regular thesis in terms of credits, publications and scientific quality. Therefore, employment for MobiDoc graduates is expected to be mostly in academia, which has been the case in Tunisia for a long time. This is due to the fact that R&D is almost nonexistent because the Tunisian economic landscape is mainly shaped by small-to-medium-sized companies. However, with the MobiDoc initiative, we may expect to see more Ph.D. students in the near future holding R&D positions in industrial and service companies.

Atidel B. Hadj-Alouane (atidel.ahadj@enit.rnu.tn) is a professor of industrial engineering and director of the OASIS Laboratory at the National Engineering School of Tunis, University of Tunis El Manar, Tunis, Tunisia.