A Simulation Comparison of Methods for New Product Location

Abstract

Four algorithms for locating an “optimal” new product in a multiattribute product space—Albers and Brockhoff's PROPOPP; Gavish, Horsky, and Srikanth's Method IV; May and Sudharshan's PRODSRCH; and GRID SEARCH—are compared in terms of the relative share of preferences the new product will capture under different simulated market environments. These environments were both ones for which the algorithms were designed as well as other “more realistic” environments. Results indicate that algorithm performance is sensitive to the number of customers or segments, and the presence of probabilistic choice, and less sensitive to the numbers of existing products. Gavish, Horsky, and Srikanth IV (GHS IV) and PROPOPP performed best under the market conditions for which they were designed and GHS IV proved quite robust under variation from these conditions. PROPOPP's performance deteriorated, however, in large sample size problems (n ≥ 200). PRODSRCH (a general purpose optimizer) was inferior under these special market conditions, but superior under other more general ones.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 77 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.