Portada documentos de trabajo

Data organisation and process design based on functional modularity for a standard production process

Doc: 01/2018

We propose to use the principles of functional modularity to cope with the essentialcomplexity of statistical production processes. Moving up in the direction of internationalstatistical production standards (GSBPM and GSIM), data organisation and processdesign under a combination of object-oriented and functional computing paradigms areproposed. The former comprises a standardised key-value pair abstract data model wherekeys are constructed by means of the structural statistical metadata of the productionsystem. The latter makes a profuse usage of the principles of functional modularity(modularity, data abstraction, hierarchy, and layering) to design production steps. Weprovide a proof of concept focusing upon an optimization approach to selective editingapplied to real survey data in standard production conditions at Statistics Spain (INE).Several R packages have been prototyped implementing these ideas. We also sharediverse aspects raising from the practicalities of the implementation.

Palabras clave / Key words: Production Architecture, Key-value Pair Data Model, Standardisation, Functional

Data organisation and process design based on functional modularity for a standard production process (Pdf 343 KB)
E. Esteban, M. Novás, S. Saldaña, D. Salgado, L.