Using simulation to optimise renting and space planning decisions

Abstract

PurposeMany modern businesses have accommodation needs which vary sharply over time. Corporate real estate (CRE) managers plan for these variations using “common sense” estimates based on average occupation levels, and these estimates are almost always wrong. This study aims to present a method where these businesses can optimise decisions on their mix of short‐ and long‐term space based on previous occupation patterns.Design/methodology/approachThe optimum accommodation mix is derived from Monte Carlo simulation, where previous work patterns are resampled to estimate future needs. The method is extended to look at the effect of rental costs, and looks at how rental decisions are affected by attitudes to risk. Extensions of the method include pricing of real estate derivatives and assessing the probability of making money from renting premises. The method is easily within the grasp of most spreadsheet users, and can be automated using a number of simple, downloadable tools. This method is suitable for organisations with fluctuating workforces, and will be of special interest to project‐based organisations.FindingsUse of “average” occupancy levels to predict the accommodation mix, while intuitive, produces worse results than simulation. While the method is transferable, the analysis must be performed using the company's specific distribution. The mix depends not only on the ratio of short‐ to long‐term rents, but also on the renter's perception of risk.Practical implicationsThe method produces provides more accuracy for accommodation planning in all “real world” cases. The technique also provides an opportunity for the CRE manager to engage with the board about the importance of business planning.Originality/valueThe study describes a simulation method widely used outside the real estate industry, and provides simple side‐bars that will help readers to create their own models using Microsoft Excel.