How ATLAS Works

Introduction

ATLAS is a collaboration of physicists, engineers, technicians, students, and support staff from around the world. It is one of the largest collaborative efforts ever assembled in science, over 200 institutes from 40 countries participate in ATLAS with around 6,000 members of whom nearly 3,000 are scientific authors (see statistics). The success of ATLAS relies on close cooperation between research teams located at CERN and at member universities, laboratories and research institutes worldwide.

ATLAS elects its leadership and has an organisational structure that enables teams to self-manage with members directly involved in decision-making processes. Scientists usually work in small groups, choosing the research areas and data that interest them most. Any output from the Collaboration is shared by all authors and is subject to rigorous review and fact-checking processes before results are made public. The success of the Collaboration depends upon individual commitment to the physics goals of the experiment and the prospect of exciting new results that can only be achieved with a complete and coherent collaborative effort.

The only way to realise such a challenging project, with the required intellectual and financial resources, and to maximise its scientific output, is through international collaboration. Collaboration resources come from investment by the funding agencies of countries participating in ATLAS, from individual institutions and charities, and from CERN itself, which is funded by member states.

Organisational Structure

The ATLAS Organisation is guided by the following principles:

  • Democracy;
  • Separation of policy-making and executive powers;
  • Minimal formal organisation;
  • Limited terms of office.

All major ATLAS Collaboration decisions concerning physics objectives and results, detector operation, changes or upgrades, and organisational matters are discussed in plenary meetings: the ATLAS Collaboration weeks and the ATLAS weekly meetings.

The Collaboration Board is the policy-making and governance body of the ATLAS Collaboration and consists of one representative for each institution, and members of the Executive Board who act as ex-officio. Institutions can be composed of single institutes or a cluster of institutes. The Chairperson of the Collaboration Board is elected by the Collaboration Board with a term of office of two years, plus two additional years serving as a deputy, and is non-renewable. The Chairperson may nominate an advisory group, the members of which will be endorsed by the Collaboration Board upon proposal of the Chairperson for the duration of the Chairperson's term.

The Executive Board directs the execution of the ATLAS project, in line with policies set by the Collaboration Board. The Executive Board is chaired by a Spokesperson, with the Technical Coordinator as Deputy Chair, and brings together the coordinators responsible for the ATLAS operation and physics data exploitation, with a view to maximise the overall performance of the experiment.

The top-level organisation chart is shown below, and full details of the ATLAS organisation are given on this webpage.

Equity, Diversity, and Inclusion (EDI) are core principles for all ATLAS activities with a dedicated EDI office that educates the Collaboration on EDI principles, promotes inclusive practices and monitors the Collaboration EDI health. More than a third of ATLAS members are early-career colleagues, with an Early Career Scientist Board representing and supporting them.

Testing,Milestones,Technology,Detectors,ATLAS

ATLAS Management

The ATLAS management team is responsible for overseeing the execution of the experiment. It is led by a Spokesperson, who is responsible for the execution of the ATLAS project, and represents the Collaboration to the LHC Committee, to the CERN Management, and to outside entities. Supporting the Spokesperson are one or two Deputy Spokespersons, whose responsibilities are defined by the Spokesperson. The team also includes a Technical Coordinator, responsible for the technical aspects of the experiment, and a Resources Coordinator, responsible for the financial and human resources of the Collaboration. Currently the management team also includes the Upgrade Coordinator who is responsible for overseeing and monitoring the ATLAS upgrades.

The ATLAS Collaboration Board elects its Spokesperson every two years, and the position is renewable once. The Spokesperson proposes deputies who are endorsed by the Collaboration Board. The Technical Coordinator and Resources Coordinator are proposed by the Spokesperson, after a call for nominations to the Collaboration. The Upgrade Coordinator is elected by the Collaboration Board. Meet the ATLAS Management.

ATLAS Detector Systems

The ATLAS detector consists of 8 different systems, each with a different role in reconstructing the LHC collisions. Each System has a Project Leader who is responsible for ensuring that the operation of the corresponding System is carried out in a way that ensures the required performance and long-term reliability. Project Leaders manage the operation and maintenance of their system, and are elected for a renewable two-year term by the representatives of the institutes participating in the given system. Some Systems have been operational since the beginning of the LHC data taking, while some will start operation during the High-Luminosity LHC phase.

Original Systems: operational since the beginning of the LHC data taking

Upgrade Systems: will become operational during the High-Luminosity LHC phase

  • Inner Tracker (ITk)
  • High-Granularity Timing Detector (HGTD)

ATLAS Activity Areas

The full range of activities from data taking to the extraction of physics results for publication is carried out by and organised into five main areas: Detector Operation, Trigger, Software and Computing, Data Preparation, and Physics. Each of these areas has a coordinator: a Run Coordinator (with a deputy) for the Detector Operation area, a Trigger Coordinator (with a deputy) for the Trigger area, two Software and Computing co-Coordinators, a Data Preparation Coordinator (with a deputy), and a Physics Coordinator (with a deputy). The coordinators are elected by the Collaboration Board with a two-year mandate serving as a deputy coordinator during the first year and as coordinator during the second year. A full organisation chart for all the activity areas can be found here.

  • Detector Operation: This area covers the detector operation for data-taking periods, which includes running the detector and the data acquisition (DAQ), recording data, checking the online data quality, etc. This work is led by the Run Coordinator.
  • Trigger: This area covers all the online event selection performance of the TDAQ system, which is organised into a Level-1 trigger and a High Level Trigger (HLT). The team defines the physics algorithms used to reduce the rate of the data selected and stored. The collection of algorithms used forms a “trigger menu”, with the menu selection having an important impact on ATLAS physics studies .
  • Software and Computing: This area covers operation of the offline computing, development and maintenance of the software, and overall technical coordination and integration of the software. The computing infrastructure includes the CERN Tier-0 centre for prompt reconstruction and over 130 computing centres worldwide.
  • Data Preparation: This area covers the work needed once the data has been recorded to prepare it for physics analysis. This includes the prompt reconstruction at the Tier-0 centre, the detector calibration and alignment, the data quality assessment, as well as the calibration constants and beam conditions used in the simulation of the ATLAS detector to match the recorded data. The measurement and accounting of the delivered and usable luminosity are also part of Data Preparation.
  • Physics: This area covers the scientific exploitation of the ATLAS experiment through the production of physics analysis results. This includes the development of sophisticated algorithms to reconstruct and calibrate the physics objects of interest (electrons, muons, jets, etc), the preparation of simulated event samples used to describe the physics processes that take place in the LHC collisions, and the development of data analyses employing statistical techniques to extract relevant physics information.
  • Outreach & Education: This area covers all aspects of communicating ATLAS research to the public and training/inspiring students of all ages. This includes the public ATLAS website (evergreen content and regular news), social media, printable material, visits and virtual visits, and open data. The group also occasionally helps with local events at CERN and supports outreach activities carried out at other ATLAS institutions.

Operational Model

Publishing scientific results using ATLAS data requires a long collaborative chain of human endeavour. As a result, all ATLAS authors are considered to be equal shareholders of the data. Consequently, everyone shares the responsibility to operate the experiment.

Operation tasks (OT) include the operation of the ATLAS detector and activities needed to prepare the data for physics analysis. Operation tasks are shared across the ATLAS institutions. The premise is based on a “fair share” model: everyone who is on the ATLAS author list (or in the process of qualifying to become an ATLAS author) should contribute to ATLAS operation tasks. In recognition of the fact that graduate students also need to take courses and write a thesis and hence are not fully available to contribute to ATLAS operation tasks, they are given a weight of 0.75 in the calculation of the number of people available to perform operation tasks.

OT are defined broadly: For each ATLAS sub-system or activity area, OT include shifts, running calibrations, data-quality monitoring, maintenance of the sub-system and its associated infrastructure, maintaining software, etc. OT do not usually include R&D into new ways to do the above tasks, and do not include physics analysis carried out in the context of the physics working groups.

The responsibility to perform OT falls on teams (i.e. institutions) or funding agencies. Early in each calendar year, the fair-share fraction is evaluated for each institution in ATLAS based on the number of authors or qualifying authors in that team as of the fall census date (9 Sept). Institutions that do significant work on the upgrade of the detector get this fair-share OT reduced in compensation, as the upgrade endeavour is essential to ensure the continuity of the experiment.

The Panel for Operation Task Sharing (POTS) oversees the fair sharing of operation tasks across the ATLAS Collaboration, and provides guidance and arbitration if required. The POTS committee is composed of the ATLAS Spokesperson and one deputy, the Collaboration Board Chairperson, the Resources Coordinator, and three members at large elected by the Collaboration Board for three years in a staggered manner (one new member is elected each year).

Physics Publication Process

ATLAS physics data analysis activities are carried out within seven combined performance, eight physics analysis groups, a statistics committee, and a simulation working group (link to org chart). Each of these groups has two co-conveners, who each normally serve for a two-year term with appointments staggered by one year. Within each group, there are subgroups, led by two co-conveners, where the work on the various topics is discussed with the analysis teams involved. Typically two contact persons are appointed for each analysis team.

ATLAS has a rigorous internal review process before physics results are made public, striving to ensure their integrity and accuracy. On average 90 journal articles are produced every year alongside summary plots and public notes explaining the details of the algorithms employed. The full list of public results can be accessed here.

The publication process is managed by the Publications Committee, which consists of 10 members endorsed by the Collaboration Board with a Chair and a Deputy elected for a two-year term. Each paper is reviewed by the relevant physics groups and by an internal editorial board before being circulated to the whole Collaboration for comments. Each analysis is presented and comments received are discussed in a Collaboration-wide paper approval meeting before results are released. The Publications Committee ensures the ATLAS publications are of high scientific and formal standard, participates in the release of conference proceedings, identifies suitable journals for ATLAS publications and ensures that the ATLAS guidelines and publication policies are followed.

Authorship of ATLAS Papers

The policy for authorship in ATLAS is guided by the principle that an ATLAS author should have made significant contributions to the experiment. Authorship is an honour and a responsibility. It is an honour to be part of a scientific endeavour at the forefront of research which is only possible in a collaboration of many devoted people with a wide spectrum of experience, skills and knowledge. At the same time authorship implies a duty to devote a significant effort to the experiment, to take responsibility as a signing author for the scientific results, and to act within the norms of scientific conduct. The full ATLAS author list is included in alphabetical order on all our publications.

ATLAS authors are expected to dedicate at least 50% of their research time to ATLAS, and may not be members of any of the other large LHC collaborations. The authorship qualification process is managed by an Authorship Committee, with members endorsed by the Collaboration Board and serving a three-year term. In order to qualify as an ATLAS author, members new to ATLAS must undertake a qualification project over a period lasting at least one year. This is typically a technical project that benefits the successful operation of ATLAS or the completion of the detector upgrades.

Speakers at Conferences

ATLAS members present about 800 talks and 200 posters at more than 100 international conferences every year. It is essential to ensure that ATLAS is well represented with high-quality presentations, as well as to guarantee a fair and equitable assignment of talks among the members of the Collaboration. To achieve this, the ATLAS Speakers Committee and Speakers Committee Advisory Board (SCAB) work together to manage the speaker assignment for presentations of ATLAS physics results. The members of the Speakers Committee and of the SCAB are endorsed by the Collaboration Board for a three- and two-year term, respectively.

The Speakers Committee provides a single contact point for conference organizers with ATLAS and manages the assignment of speakers and the review of their talks given on behalf of the ATLAS Collaboration. Talks covering technical aspects of detectors (including future plans), Outreach and Education, and EDI are handled by the various Subsystem Speakers committees. Personal invitations for ATLAS physics and overview talks at international conferences and workshops are handled by the Speakers Committee and ATLAS members are required to obtain the approval of the Speakers Committee before accepting invitations.

The primary functions of the SCAB are (1) to define the priority (called “SCAB rank”) of each ATLAS author to be assigned as a speaker at international conferences and (2) to provide oversight to the Collaboration regarding both the guidelines used in the selection of ATLAS speakers and the equitable distribution of talks among the members of the Collaboration. The SCAB rank takes into account several parameters, including the involvement of each ATLAS author in different areas of the experiment, as well as nominations from management, project leaders, activity coordinators and institute representatives. Nominations are given for both physics-related efforts and essential contributions to the experiment operation and upgrade activities. Other information, such as professional advancement, career stage, and time since the last presentation, is also considered. To obtain a SCAB rank ATLAS members need to deliver a minimum OT contribution and receive at least two nominations.

ATLAS Funding Model

The ATLAS experiment is supported by funding agencies from the 40 countries of the institutes participating in the Collaboration. The basis of the funding model was established in a memorandum of understanding (MoU) signed in 1998 (see here) to cover the costs of the experiment construction and a second MoU in 2002 (see here) regarding the maintenance and operation (M&O) of the experiment. The M&O costs are divided into different categories, as described in article 7 of the second MoU, including its sharing in article 9.

A Resources Review Board (RRB) has been constituted which comprises the representatives of all ATLAS Funding Agencies and the managements of CERN and the ATLAS Collaboration. It is chaired by the CERN Director of Research and Computing. The management of the Collaboration reports twice a year to the RRB on technical, managerial, financial and administrative matters, and on the composition of the Collaboration. The contact between the funding agencies and ATLAS management is facilitated via National Contact Physicists.

Working with ATLAS

There are several types of ATLAS membership available from full institution members represented in the Collaboration Board to technical associate institutes as described in this page. Typically, new institutes join ATLAS as associate members of an existing institution before applying for full membership.

Furthermore, it is possible for non-ATLAS members to collaborate with ATLAS as short term associates, or interact with ATLAS physicists on Monte Carlo generators and related tools. Full details are given here.

Open Access, Data and Software

The following statement was approved by the ATLAS Collaboration Board in 2007:

We, the ATLAS Collaboration, strongly encourage the usage of electronic publishing methods for ATLAS publications and support the principles of Open Access Publishing, which includes granting free access of our ATLAS publications to all. Furthermore, we encourage all ATLAS members to publish papers in easily accessible journals, following the principles of the Open Access Paradigm.

ATLAS has fully supported the principle of open access in its publication policy as described in the CERN Open data policy for LHC experiments: CERN-OPEN-2020-013. Further information on the ATLAS open data initiative can be found at this website. The software written by ATLAS is publicly available and is distributed using an Apache 2 licence agreement.

ATLAS Awards

The ATLAS Collaboration recognises exceptional contributions to the experiment through two distinct awards.

  • ATLAS Doctoral Awards (named ATLAS Thesis Awards prior to 2026): Students contribute strongly and critically to all areas of the experiment while learning valuable skills for their degrees. Awarded annually since 2010, these recognise students’ outstanding contributions in the context of their PhD work. Candidates are nominated by their supervisors or home institute scientists. The selection is done by the Collaboration Board Advisory Group based on scientific quality, overall impact, breadth of the contribution, and clarity and pedagogy of the thesis presentation. The number of students awarded every year varies between 3 and 8.
  • ATLAS Outstanding Achievements Awards: Introduced in 2014, these (bi)annual awards recognise exceptional accomplishments by individuals or teams during a given period in all areas of the ATLAS experiment, excluding physics analysis. This includes contributions to technical coordination, detector maintenance, operations and upgrade, data preparation, trigger, combined performance, software and computing activities. Any active ATLAS member or team during the reference period is eligible for the award. Nominations are open to all active Collaboration members. Nominations come from across the Collaboration. The winners are selected by a committee, involving experts’ evaluation in each area and cross-evaluations done by the Collaboration Board Chair Advisory Group. The number of achievements awarded every year varies between 5 and 13.