Operator-Paced Line Speed, and Computerized Layout Programs (CRAFT/CORELAP): 23-Ind-A3 Facilities Planning, May 2015, Q4

Question 4: Operator-Paced Line Speed, and Computerized Layout Programs (CRAFT/CORELAP) (20 marks: i–8, ii–7, iii–5)

Question text not reproduced: the examination questions are © Engineers and Geoscientists BC. Open the official past paper (linked at the top of this page) to read the question, then follow the worked solution below.

(i) Station Speed for a Paced Line Under Normally Distributed Operator Performance

Check — assumption

The source gives the performance range (60%–140%) but not the standard deviation directly. Following the standard treatment of this problem type, the stated range is taken to span ±3 standard deviations from the mean (the usual textbook convention for "range of performance" data of this kind) — i.e. $140-100=100-60=3\sigma$. This is flagged explicitly because the exam does not state the sigma-multiple outright.

Given. Operator performance rating $\sim N(\mu=100\%,\ \sigma)$; stated range 60%–140% (taken as $\mu\pm3\sigma$); $Z=1.04$ for the 85th/15th percentile.

Find. (a) the station speed (as a % of average) the paced line must be set to if it is to accommodate the operator at the 15th percentile of the performance distribution (i.e. the operator whom 85% of the workforce out-paces — described in the question as "85% of average"); (b) the gain in station speed if the line is instead decoupled and set for the average (100%) operator.

Approach. Derive $\sigma$ from the stated range, then use the given $Z=1.04$ to find the performance rating at the 15th percentile (the operator a synchronous/paced line must be set for so that only the slowest 15% of the workforce cannot keep pace); compare that rating to the 100% average rating to get the decoupled-line gain.

Standard deviation of the performance distribution. With the range taken as $\mu\pm3\sigma$: $3\sigma=140-100=100-60=40\%\Rightarrow\sigma=\boxed{13.33\%}$.
Performance rating at the 15th percentile (the paced-line setting). A synchronous line paced for the AVERAGE (100%) operator would leave roughly half the workforce unable to keep up, so it is instead paced for an operator below average — specifically the operator at the 15th percentile, so that 85% of the workforce can sustain the line's speed. Using the lower-tail Z-value ($Z=-1.04$): $$X_{15}=\mu+Z\sigma=100+(-1.04)(13.33)=100-13.87=\boxed{86.13\%\text{ of average pace}}$$ This is the station speed the line must be set to.
(b) Gain in station speed if decoupled and set for the average operator. Decoupling (Question 5(i)) removes the need to pace every station to the slowest-accommodated worker, so the line can instead run at the average (100%) rate: $$\text{Gain}=100\%-86.13\%=\boxed{13.87\text{ percentage points}}$$ expressed relative to the paced-line speed, this is a $\dfrac{13.87}{86.13}\times100\%=\boxed{16.1\%}$ increase in line speed.

Quantity	Result
$\sigma$ of the performance distribution	13.33%
(a) Paced-line station speed (15th-percentile operator)	86.13% of average
(b) Gain from decoupling & setting for the average operator	13.87 points (16.1% faster)

(ii)(a) CRAFT

CRAFT (Computerized Relative Allocation of Facilities Technique) is an improvement-type (also called improvement/heuristic) computerized layout algorithm. It starts from a user-supplied INITIAL layout (an existing layout, or an initial guess) together with a from–to flow matrix and a cost/distance matrix, then evaluates pairwise (2-way or 3-way) interchanges of department locations, accepting an interchange whenever it reduces total material-handling cost ($\sum f_{ij}c_{ij}d_{ij}$), and continues until no further interchange improves the objective (a local-search/steepest-descent heuristic). Because it only swaps departments of roughly compatible area (it does not freely reshape department boundaries beyond swapping), CRAFT can produce impractical, non-rectangular department shapes that a human planner must clean up; it also, being a local-search method, can converge to a local rather than global optimum, so the quality of its result is sensitive to the quality of the starting layout supplied.

(ii)(b) CORELAP

CORELAP (Computerized Relationship Layout Planning) is a construction-type algorithm — it builds a layout from scratch rather than improving a given one. Its input is the qualitative activity-relationship (REL) chart (the A-E-I-O-U-X closeness ratings and reason codes), not a quantitative from–to flow matrix. CORELAP computes a Total Closeness Rating (TCR) for each department (a weighted sum of its relationship codes with every other department) and places departments into the layout one at a time in decreasing order of TCR, each new department positioned to best satisfy its closeness relationships with the departments already placed. Because it works from qualitative relationship data, CORELAP is well suited to situations where the desired adjacencies are known but numeric flow volumes are not (or are not the dominant siting criterion, e.g., where supervision, safety or shared-service adjacency matters as much as flow); like CRAFT, though, it is a heuristic and does not guarantee a globally optimal layout.

(iii) Basic Requirements of Computerized Layout Programs for Multiple Items

To lay out a facility handling multiple products/items, a computerized layout program needs, at minimum: (1) a defined set of departments/activities and their space requirements (Question 3(i)/(ii)); (2) a quantitative interaction measure between every pair of departments — either a from–to flow/cost matrix (as CRAFT requires) or a qualitative relationship chart (as CORELAP requires) — that aggregates the flow or closeness need across ALL items/products handled, not just one; (3) a building/site outline or area constraint within which the departments must fit; (4) an objective function (typically minimize total material handling cost, or maximize weighted closeness satisfaction) that the algorithm can evaluate and improve against; (5) a starting layout (for improvement-type algorithms like CRAFT) or a placement rule (for construction-type algorithms like CORELAP); and (6) the ability to respect physical/fixed-location constraints (fixed-position equipment, building columns, docks) that cannot be relocated by the algorithm. Because these programs optimize a single aggregated objective across the full department set, when multiple items/products are handled their individual flow volumes (or relationship needs) must first be combined into ONE composite interaction matrix — this aggregation step, and the choice of relative weighting between products, is itself a modeling decision that materially affects the resulting layout.

Concept & Theory

Topic. Operator-paced (synchronous) line speed under normally distributed individual performance; computerized layout algorithms (improvement-type CRAFT vs. construction-type CORELAP) and their input/data requirements.

Key relations. Paced-line speed setting: $X_p=\mu+Z_p\sigma$ for the target percentile $p$; CRAFT objective $\min\sum f_{ij}c_{ij}d_{ij}$ via pairwise interchange from an initial layout; CORELAP placement via descending Total Closeness Rating from a REL chart.

Why this works. A synchronous line forces every operator to the SAME speed, so setting it at the population average would stall roughly half the workforce — setting it at a lower percentile (here the 15th) trades some line speed for the fraction of the workforce that can sustain it, a trade only removed by decoupling (buffers/parallel stations, Question 5(i)). CRAFT and CORELAP differ because they start from different information: CRAFT needs quantitative flow data and an initial layout to improve; CORELAP needs only qualitative closeness ratings and builds outward from the highest-priority department.

Common pitfall. Reading "85% of average" literally as the number 85 rather than as the operator positioned at the 15th percentile of the distribution (whose precise rating, from the given Z-table value, is 86.13%, not exactly 85%); confusing CRAFT (needs quantitative flow + starting layout, improvement-type) with CORELAP (needs only qualitative REL data, construction-type).

Source. Niebel & Freivalds, Methods, Standards, and Work Design, Ch. 16 (paced lines and normal-distribution performance variability); Tompkins, White, Bozer & Tanchoco, Facilities Planning, Ch. 6 (Layout Planning Models and Design Algorithms — CRAFT, CORELAP and related programs).