AUTOMATED LAYOUT DESIGN PROGRAM (ALDEP)

Now we will examine Automated Layout Design Program (ALDEP).

ALDEP is basically a construction algorithm but it can also be used to evaluate two layouts. The algorithm uses basic data on facilities and builds a layout by successively placing the layout using relationship information between the departments. The basic inputs to ALDEP are:

1. Length and width of facility.2. Area of each department.3. Minimum closeness preference (MCP) value.4. Sweep width.5. Relationship chart showing the closeness rating.6. Location and size of any restricted area.

The procedures adopted for using ALDEP are:

Step 1: Input the following:

1. Length and width of facility.2. Area of each department.3. Minimum closeness preference (MCP) value.4. Sweep width.5. Relationship chart showing the closeness rating.6. Location and size of restricted area.

Step 2: One department is selected randomly and placed in the layout.

Step 3: In this step, the algorithm uses minimum closeness required between departments for the selection of departments to be placed with an earlier placed department. Select the department having maximum closeness rating. If there is no department having minimum closeness preference then any department that remains to be placed is selected.

Step 4: If all the departments are placed in the layout, go to step 5. Else, go to step 3.

Step 5: Compute the total score of the layout.

Step 6: If the total score required is the acceptable score, then go to step 7, else go to step 2.

Step 7: Print the current layout and the corresponding score.

The following example illustrate the approach in ALDEP, heuristics.

Example

Develop a layout for the following problem.

1. Layout and area requirements are shown in Table below.Table: Layout Requirements

Department Area (sq. ft) Number of unit squares1 1200 302 800 203 600 154 1200 305 800 206 1200 307 1200 30

Total 7000 175Assume one square in the layout to be equal to 40 sq. ft.

Number of unit squares for a department = dept. area in sq. ft/area per square.

Let the size of layout be 15 × 12, and the sweep width be 2 (this means that we will fill 2 columns simultaneously).

The relationship chart for the example is as follows:

Department

1 2 3 4 5 6 7

1 E O I O U U2 E U E I I U3 O U U U O U4 I E U I U U5 O I U I A I6 U I O U A E7 U U U U I E

In ALDEP the closeness rating uses the following notations with the following values:

Chosen closeness rating

Absolutely necessary A 64Especially important E 16Important 1 4Ordinary O 1Unimportant U 0Undesirable X –1024

Randomly select the first department in the layout. Place the first department in the upper left corner and extend it downward. Width of the extension is determined by the sweep width. The next department begins where the previous department ended and follows the serpentine sweep pattern (shown in figure below.)

Let department 2 be selected. Number of unit squares in department 1 be 20. Now, 20 square units are filled in 15 × 12 grids as shown in figure below.

Since the minimum closeness between departments required for selection of departments is I = 4, scan the relationship chart randomly to find the departments having closeness rating of 4 or greater with department 2. For the above case closeness rating for the pair (1–2) = 16, (2–4) = 16, (2–-5) = 4 and (2–6) = 4.

Select any department say, department 1. Place department 1 in the layout in a serpentine pattern as shown in the figure below.

Repeat the above procedure to get the final layout as shown in figure below.

Sweep pattern

Placing department 2 in Layout

Placing department 1 in the layout

After the final layout is obtained, the score is calculated. The score is the sum of the closeness ratings of all the neighbouring departments as shown in table below.

Table: Closeness rating

Department pair Closeness rating(1, 2) 16(1, 4) 4(1, 5) 1(4, 5) 4(5, 6) 64(4, 6) 0(5, 7) 4(6, 7) 16(7, 3) 1(6, 3) 0Total 110

From the above, the layout score is 2 × 110 = 220. A further iteration should be carried out to check if a better score can be achieved.

Features of CRAFT

The major features of CRAFT are as listed below:

Attempts to minimize transportation cost, where transportation cost = flow × distance × unit cost.

Required the assumptions that: (1) move costs are independent of the equipment utilization and (2) move costs are linearly related to the length of the move.

Distance matrix used in the rectilinear distance between department centroids.

CRAFT being a path-oriented method, the final layout is dependent on the initial layout. Therefore, a number of different initial layouts should be used as input to the CRAFT layout. CRAFT allows the use of dummy departments to represent fixed areas in the layout. CRAFT input requirements are:

1. Initial layout.2. Flow data.3. Cost per unit distance.4. Total number of departments.5. Fixed departments and their location.6. Area of departments.

The procedures adopted for using CRAFT are:

Final Layout

1. Determine department centroids.2. Calculate rectilinear distance between centroids.3. Calculate transportation cost for the layout.4. Consider department sharing a common border.5. Determine transportation cost of each departmental interchange.6. Select and implement the departmental interchange that offers the greatest reduction in

transportation cost.7. Repeat the procedure for the new layout until no interchange is able to reduce the

transportation cost.

Major disadvantages of using CRAFT

CRAFT, inspite of its popularity has some major drawbacks.

Because the basis is the cost of materials handling, only production departments are considered. No service departments are considered.

An initial idea of the layout is required. Therefore, the technique only applies to the modification of an existing layout or new layouts where the outline shape is known.

The distances between the departments is taken as straight lines whereas in practice movement is usually rectangular along orthogonal lines.

To illustrate the approach to develop a layout, in CRAFT heuristics consider the initial layout and flow data as in the following example. Assume that the cost per unit transfer to be 1.

Example

Consider the following layout problem with unit cost matrix. Use CRAFT algorithm to obtain layout. The initial layout is shown in figure below and the flow matrix in table ? ? ? ?

Table: Flow Matrix

Department A B C DA 30 25 45B 20 15 20C 10 20 10

Initial Layout

D 100 10 5Centroids of all the departments in the initial layout are calculated and presented as given below:

(XA, YA) = 10.5, 3.5

(XB, YB) = 10.5, 10.5

(XC, YC) = 3.5, 10.5

(XD, YD) = 3.5, 3.5

Using the rectilinear distance we draw the distance matrix as shown in table below.

Table: Distance Matrix

Department

A B C D

A 0 7 14 7B 7 0 7 14C 14 7 0 7D 7 14 7 0

Total material handling cost is calculated as follows:

Total cost = flow × distance × unit cost

Table: Total cost matrix

Department

A B C D Cost

A 0 210 350 315 875B 140 0 105 280 525C 140 140 0 70 350D 700 140 35 0 875

Total Cost 2625Consider various departmental interchange for improvement. Departmental interchange is possible for departments having common boundary or equal area. The possible department interchanges are as shown in table below.

Table: Departmental interchanges

Department pair ReasonA-B Common border and equal areaA-C Equal areaA-D Common border and equal areaB-C Common border and equal areaB-D Equal areaC-D Common border and equal area

For the purpose of calculating material handling cost, interchange would mean change in the centroid. In the same way as we calculated the total cost for the initial layout, we calculate the total cost for each of the possible interchanges, and select the layout that gives the least total cost.